Dynamic range control for a wide variety of playback environments

Mutual reference to related applications This application applies to US Provisional Patent Application No. 61 / 877,230 filed on September 12, 2013, US Provisional Patent Application No. 61 / 891,324 filed on October 15, 2013, and 2014. It claims the priority of US Provisional Patent Application No. 61 / 938,043 filed on February 10. The content of each application is incorporated herein by reference in its entirety.

Technology The present invention relates generally to the processing of audio signals, and more specifically to applying dynamic range control and other types of audio processing operations to audio signals in any of a wide variety of playback environments. Regarding techniques that can be used.

The growing popularity of media consumer devices has created new opportunities and challenges for creators and distributors of media content for playback on such devices, or for designers and manufacturers of such devices. Many consumer devices can play a wide range of media content types and formats, including those often associated with high quality, wide bandwidth and wide dynamic range audio content for HDTV, Blu-ray or DVD. .. Media processors can use this type of audio content to play on their own internal acoustic transducers or on external transducers such as headphones. However, media processors are generally unable to play this content with consistent loudness and comprehension across a variety of media formats and content types.

The approaches described in this section could be pursued, but are not necessarily previously conceived or pursued approaches. Therefore, unless otherwise noted, none of the approaches described in this section should be assumed to qualify for prior art just because they are included in this section. Similarly, unless otherwise noted, issues identified with respect to one or more approaches should not be assumed to have been recognized in any prior art based on this section.

The present invention is shown in the accompanying drawings as an example, but not a limitation. In the drawings, similar reference numerals refer to similar elements.<figref num="1A">It is a figure which shows an exemplary audio decoder.</figref><figref num="1B">It is a figure which shows an exemplary audio encoder.</figref><figref num="2A">It is a figure which shows an exemplary dynamic range compression curve.</figref><figref num="2B">It is a figure which shows an exemplary dynamic range compression curve.</figref><figref num="3">It is a figure which shows the exemplary processing logic of the combination DRC and the limit gain determination / calculation.</figref><figref num="4">It is a figure which shows the example differential coding of gain.</figref><figref num="5">FIG. 5 illustrates an exemplary codec system with an audio encoder and an audio decoder.</figref><figref num="5A">It is a figure which shows an exemplary audio decoder.</figref><figref num="6A">It is a figure which shows the exemplary process flow.</figref><figref num="6B">It is a figure which shows the exemplary process flow.</figref><figref num="6C">It is a figure which shows the exemplary process flow.</figref><figref num="6D">It is a figure which shows the exemplary process flow.</figref><figref num="7">FIG. 5 illustrates an exemplary hardware platform on which the computers or computing devices described in this article can be implemented.</figref>

Illustrative embodiments of applying dynamic range control and other types of audio processing operations to audio signals in any of a wide variety of playback environments are described herein. In the following description, for the purposes of explanation, a number of individual details are provided for the purposes of explanation to provide a thorough understanding of the present invention. However, it will be clear that the present invention can be practiced without such individual details. On the other hand, in order to avoid unnecessarily concealing, haze, or burying the present invention, well-known structures and devices are not described in exhaustive detail.

Illustrative embodiments are described herein according to the outline below. 1. General overview 2. Dynamic range control 3. Audio decoder 4. Audio encoder 5. Dynamic range compression curve 6. DRC gain, gain limitation and gain smoothing 7. Input smoothing and gain smoothing 8. Multiple frequencies DRC over band 9. Volume adjustment in loudness region 10. Gain profile with differential gain 11. Additional gain-related behavior 12. Specific and broadband (or broadband) loudness levels 13. For individual subsets of channels Individual gains of 14. Auditory scene analysis 15. Loudness level transitions 16. Resets 17. Gains provided by encoders 18. Illustrative system and process flow 19. Implementation mechanism-Hardware overview 20. Equivalents , Extensions, alternatives and more.

<u style="single">&lt;1. General overview&gt;</u> This overview presents a basic description of some aspects of embodiments of the present invention. It should be noted that this overview is not a comprehensive and exhaustive summary of the aspects of the embodiment. Further, this overview is understood to identify some particularly significant aspect or element of the embodiment, and is generally understood to define some scope of the invention, in particular the embodiment. It should also be noted that it was not intended. This overview merely presents some concepts related to the exemplary embodiment in a condensed and simplified form, and merely a concept to a more detailed description of the subsequent exemplary embodiments. It should be understood as a general introduction. Although separate embodiments are discussed herein, it should be noted that any combination of embodiments and / or partial embodiments discussed herein may be combined to form further embodiments.

In some approaches, the encoder assumes that the audio content is encoded for a particular environment for the purpose of dynamic range control, for that particular environment, for dynamic range control, etc. Determine audio processing parameters such as gain in. The gain determined by the encoder under these approaches is typically smoothed over some time interval, with some time constant (eg, in an exponential decay function). In addition, the gain determined by the encoder under these approaches may be built in to limit the gain to ensure that the signal does not exceed the clipping level for the expected environment. Thus, the gain encoded by the encoder in the audio signal along with the audio information under these approaches is the result of many different effects and is irreversible. Decoders that receive gain under these approaches have which part of the gain is for dynamic range control, which part of the gain is for gain smoothing, which part of the gain is for gain limitation, etc. Will not be distinguishable.

Under the techniques described in this paper, audio encoders do not assume that a particular playback environment in an audio decoder will only be supported. In one embodiment, the audio encoder transmits an encoded audio signal with audio content from which the correct loudness level can be determined (eg, without clipping). The audio encoder may also send one or more dynamic range compression curves to the audio decoder. Either of the one or more dynamic range compression curves may be standards-based, proprietary, customized, content provider specific, or the like. Even if the reference loudness level, attack time, release time, etc. are transmitted by the audio encoder as part of the one or more dynamic range compression curves or in connection with the one or more dynamic range compression curves. Good.

In some embodiments, the audio encoder implements an auditory scene analysis (ASA) technique and uses the ASA technique to detect auditory events in audio content, using the detected auditory events. Send one or more ASA parameters to describe to the audio decoder.

In some embodiments, the audio encoder detects a reset event in the audio content and sets the index of the reset event downstream along with the audio content in a time-synchronous manner, such as an audio decoder. It can also be configured to send to the device of.

In some embodiments, the audio encoder is one or more gains (such as DRC gains) for individual parts of the audio content (eg audio data blocks, audio data frames, etc.). The set can be calculated and the set of gains can be configured to be encoded in the encoded audio signal together with the individual parts of the audio content. In some embodiments, the set of gains produced by the audio encoder corresponds to one or more gain profiles (eg, as shown in Table 1). In some embodiments, Huffman coding, differential coding, etc. encode the set of gains into components, subdivisions, etc. of various audio data frames, or the set of gains from the components, subdivisions, etc. It may be used to read. These components, subdivisions, etc. are sometimes referred to as subframes of audio data frames. Sets with different gains may correspond to sets with different subframes. Each set of gains or each set of subframes may have two or more temporal components (eg, subframes, etc.). In some embodiments, the bitstream formatters in the audio encoders described herein use one or more for loops to use one or more sets of gains as differential data codes for audio. May be written together in one or more sets of subframes in a data frame. Correspondingly, the bitstream parser in the audio decoder described herein is one of the gains encoded as the difference data code from the one or more sets of subframes in the audio data frame. You may read either one or more sets.

In some embodiments, the audio encoder determines the dialog loudness level in the audio content to be encoded in the encoded audio signal, and sets the dialog loudness level along with the audio content. Send to the decoder.

In some embodiments, the audio encoder sends a downstream receiving audio decoder a default dynamic compression curve for the default gain profile in the playback environment or scenario. In some embodiments, the audio encoder assumes that the downstream receiving audio decoder uses the default dynamic compression curve for the default gain profile in the playback environment or scenario. In some embodiments, the audio encoder is downstream as to which of one or more dynamic compression curves defined in the downstream receiving audio decoder should be used in the playback environment or scenario. Send an index to the receiving audio decoder. In some embodiments, for each of one or more non-default gain profiles, the audio encoder corresponds to that non-default gain profile as part of the metadata carried by the encoded audio signal ( Send a dynamic compression curve (for example, non-default). The techniques described in this paper allow multiple sets of differential gains associated with the default compression curve to be generated by the upstream encoder and sent to the downstream decoder. This means designing the DRC compressor in the decoder (for example, the process of calculating gain based on the compression curve and the smoothing operation) while keeping the required bit rate relatively low compared to transmitting the full gain value. Allows a great deal of freedom in. For illustration purposes only, the default profile or default DRC curve was mentioned as being able to specifically calculate the differential gain for the non-default profile or non-default DRC curve in relation to it. However, this is just for illustration purposes and there is no strict need to distinguish between default and non-default profiles (for example, in media data streams). In various embodiments, all other devices This is because the rofile can be the differential gain relative to the same specific compression curve (eg, "default"). In the usage in this paper, the "gain profile" is sometimes referred to as the DRC mode as the operating mode of the compressor that performs the DRC operation. In some embodiments, the DRC mode relates to the specific type of playback device (AVR or TV or tablet) and / or the environment (noisy or quiet or midnight). Each DRC mode can be associated with a gain profile. The gain profile may be represented by definition data, based on which the compressor performs a DRC operation. In some embodiments, the gain profile can be a DRC curve (possibly parameterized) and a time constant used in the DRC operation. In some embodiments, the gain profile can be a set of DRC gains as an output of DRC operation in response to an audio signal. Profiles in different DRC modes may accommodate different amounts of compression.

In some embodiments, the audio encoder is a set of default gains (eg, full DRC and non-DRC, full DRC, etc.) for audio content based on a default dynamic range compression curve that corresponds to the default gain profile. To determine the set of non-default gains (eg, full DRC and non-DRC, full DRC, etc.) for the same audio content for each of one or more non-default gain profiles. The audio encoder then sets the default gain profile (eg, full DRC and non-DRC, full DRC, etc.) and the non-default gain profile (eg, full DRC and non-DRC, full DRC, etc.). The difference in gain from the set of gains can be determined and the difference in gains can be included in the set of differential gains. Instead of sending a dynamic compression curve (for example, non-default) for a non-default profile related to a non-default playback environment or scenario, the audio encoder puts a set of differential gains instead of the non-default dynamic compression curve or In addition, it can be sent as part of the metadata carried by the encoded audio signal.

The set of differential gains may be smaller than the set of non-default (eg, full DRC and non-DRC, full DRC, etc.) gains. Thus, transmitting differential gain rather than non-differential (eg full DRC and non-DRC, full DRC, etc.) gain is directly transmitting non-differential (eg, full DRC and non-DRC, full DRC, etc.) gain. In comparison, the required bit rate may be lower.

The audio decoders that receive the encoded audio signals described herein may be provided by different manufacturers and are implemented with different components and designs. The audio decoder may have been released to the end user at different times, or may have been updated with different versions of hardware, software, or firmware. As a result, those audio decoders may have different audio processing capabilities. In some embodiments, a large number of audio decoders may have the ability to support a limited set of gain profiles, such as default gain profiles defined by standards, proprietary requirements, and so on. Many audio decoders may be configured with the ability to perform relevant gain generation operations to generate gain for the default gain profile based on a default dynamic range compression curve representing the default gain profile. .. Transmitting the default dynamic range compression curve for the default gain profile in the audio signal can be more efficient than transmitting the gain generated / calculated for the default gain profile in the audio signal.

On the other hand, for non-default gain profiles, the audio encoder can pre-generate the differential gain with reference to a particular default dynamic range compression curve corresponding to a particular default gain profile. In response to receiving the differential gain in the audio signal generated by the audio encoder, the audio decoder generates and receives the default gain based on the default dynamic range compression curve received in the audio signal. The received audio content is combined, for example, by combining the differential gain generated and the default gain generated with the non-default gain for the non-default gain profile, and applying the non-default gain to the audio content decoded from the audio signal. Can be rendered. In some embodiments, the non-default gain profile may be used to guarantee a limitation of the default dynamic range compression curve.

The techniques described in this article can be used to provide flexible support for new gain profiles, features or improvements. In some embodiments, at least one gain profile, whether default or non-default, cannot be easily represented using a dynamic range compression curve. In some embodiments, the at least one gain profile may be specific to a particular audio content (eg, a particular movie). The representation of a non-default gain profile (eg, a parameterized DRC curve) requires that more parameters, smoothing constants, etc. that can be carried in the encoded audio signal be transmitted in the encoded audio signal. There can be. In some embodiments, at least one gain profile may be specific to a particular audio content provider (eg, a particular studio).

As such, the audio encoders described in this article can take the lead in supporting new gain profiles. It is by implementing a gain generation operation for the new gain profile and a gain generation operation for the default gain profile to which the new gain profile is involved. The downstream receiving audio decoder does not need to perform a gain generation operation for the new gain profile. Rather, the audio decoder supports the new gain profile by taking advantage of the non-default differential gain generated by the audio encoder without the audio decoder performing gain generation operations for the new gain profile. it can.

In some embodiments, one or more dynamic range compression curves (eg, default) and differential gain (eg, non-default) in the profile-related metadata encoded in the encoded audio signal. According to the respective gain profiles to which one or more sets correspond to, for example, the one or more dynamic range compression curves (such as default) and one or more sets of differential gains (such as non-default). Structured, indexed, etc. In some embodiments, the relationship between the set of non-default differential gains and the default dynamic range compression curve may be indicated in the profile relationship metadata. This can be particularly useful when there are two or more default dynamic range compression curves in the metadata, or even if they are not in the metadata, they are defined in the downstream decoder. Based on the relationships indicated in the profile relationship metadata, the receiving audio decoder can determine which default dynamic range compression curve should be used to generate the set of default gains. The generated gain can then be combined with the received set of non-default differential gains to generate non-default gains, for example to compensate for the limitations of the default dynamic range compression curve.

The techniques described in this article may have been performed by an audio decoder by an upstream device such as an audio encoder, assuming a hypothetical playback environment, scenario, etc. in a hypothetical audio decoder (eg,). Does not require being locked in with audio processing (such as irreversible). The decoders described in this article distinguish between different loudness levels present, for example in audio content, minimize the loss of audio perceptual quality at or near the boundary loudness level, and are a channel or subset of channels. It may be configured to customize audio processing behavior based on a particular playback scenario, such as to maintain spatial balance between them.

An audio decoder that receives an encoded audio signal with a dynamic range compression curve, reference loudness level, attack time, release time, etc. determines the particular playback environment used in the decoder and that particular playback environment. A specific compression curve with the corresponding reference loudness level can be selected.

The decoder calculates / determines the loudness level in individual parts of the audio content extracted from the encoded audio signal (eg, audio data blocks, audio data frames, etc.), or the audio encoder determines the loudness. -If the level is calculated and provided in the encoded audio signal, the loudness level in each part of the audio content can be obtained. Loudness level in individual parts of audio content, loudness level in previous parts of audio content, loudness level in subsequent parts of audio content, if available, said specific compression curve, said specific Based on one or more of the specific profiles related to the playback environment or scenario, the decoder provides audio processing parameters such as gain (DRC gain) for dynamic range control, attack time, release time, etc. decide. Audio processing parameters can also include adjustments to align the dialog loudness level with a particular reference loudness level (which may be user adjustable) for a particular playback environment.

With the audio processing parameters, the decoder applies audio processing operations including dynamic range control (eg, multi-channel, multi-band, etc.), dialog level adjustment, and the like. The audio processing operations performed by the decoder also include gain smoothing based on attack and release times provided as part of or in connection with the selected dynamic range compression curve, gain limiting to prevent clipping, etc. May include, but is not limited to. Different audio processing operations may be performed with different (eg, adjustable, threshold dependent, controllable) time constants. For example, gain limits to prevent clipping may be applied to individual audio data blocks, individual audio data frames, etc. with relatively short time constants (eg momentary, about 5.3 ms, etc.). Good.

In some embodiments, the decoder extracts ASA parameters from the metadata in the encoded audio signal, such as the time position of the auditory event boundary, the time-dependent value of the event certainty indicator, and the extracted ASA. Control the rate of gain smoothing in the auditory event based on parameters (for example, use a short time constant for attack at the auditory event boundary, use a long time constant to slow down gain smoothing in the auditory event, etc.) Can be configured.

In some embodiments, the decoder also maintains a histogram of the instantaneous loudness level for a time interval or window, and the histogram is used to gain in loudness level transitions, such as between programs and between programs and commercials. It is used to control the rate of change, for example by modifying the time constant.

In some embodiments, the decoder supports more than one speaker configuration (eg, portable mode with speakers, portable mode with headphones, stereo mode, multi-channel mode, etc.). The decoder may be configured to maintain the same loudness level between two different speaker configurations (eg, between stereo mode and multichannel mode), for example when playing the same audio content. Good. The audio decoder may down-mix one or more of the multi-channel audio content received from the encoded audio signal for a reference speaker configuration to a particular speaker configuration in the audio decoder. You may use the mix formula. The multi-channel audio content is encoded for the reference speaker configuration.

In some embodiments, automatic gain control (AGC) may be disabled in the audio decoders described herein.

In some embodiments, audiovisual devices, flat panel TVs, handheld devices, game consoles, televisions, home theater systems, tablets, mobile devices, laptop computers, netbook computers, cellular wireless phones, e-book readers. , Point of sale terminals, desktop computers, computer workstations, computer kiosks, various other types of terminals and media processing units, etc., but are not limited to being part of a media processing system.

Various modifications to the preferred practices and general principles and features described in this article will soon become apparent to those skilled in the art. Accordingly, this disclosure is not intended to be limited to the embodiments shown and is given the broadest scope consistent with the principles and features described herein.

<u style="single">&lt;2. Dynamic range control&gt;</u> Without customized dynamic range control, input audio information (eg, PCM samples, time-frequency samples in a QMF matrix, etc.) is often played back in a particular playback environment of the playback device (ie, physical and / or mechanical playback of the device). Plays at an inappropriate loudness level (including limits). This is because the particular playback environment of the playback device can differ from the playback environment targeted when the encoded audio content was encoded in the encoding device.

The techniques described in this article can be used to support dynamic range control of a wide variety of audio content customized to any of a wide variety of playback environments while maintaining the perceptual quality of the audio content. ..

Dynamic range control (DRC) is the time it takes to change the loudness level input dynamic range of audio content to an output dynamic range that is different from the input dynamic range (eg, compress, cut, expand, boost, etc.). Dependent audio processing operation. For example, in a dynamic range control scenario, soft sounds may be mapped to higher loudness levels (eg boost) and loud sounds may be mapped to lower loudness values (eg cut). As a result, in the loudness region, in this example, the loudness level output range is smaller than the loudness level input range. However, in some embodiments, the dynamic range control may be reversible so that the original range is restored. For example, each unique original loudness level maps to a unique output loudness level where the mapped loudness level in the output dynamic range mapped from the original loudness level is less than or equal to the clipping level. As long as it is done, a decompression operation can be performed to restore the original range.

The DRC techniques described in this article can be used to provide a better listening experience in certain playback environments or situations. For example, a soft sound in a noisy environment can be masked by noise that makes the soft sound inaudible. Conversely, in some situations, such as a noisy neighbor, loud sounds may not be desired. Many devices, typically with small Scherrer loudspeakers, are unable to reproduce sound at high output levels. In some cases, lower signal levels can be reproduced below the human auditory threshold. The DRC technique maps input loudness levels to output loudness levels based on DRC gains found using dynamic range compression curves (eg scaling factors that scale audio amplitude, boost ratios, cut ratios, etc.). Can do that.

A dynamic range compression curve is a function that maps individual input loudness levels (such as non-dialog sounds) determined from individual audio data frames to individual gains or gains for dynamic range control. (For example, lookup table, curve, multi-segment segmented straight line, etc.). Each of the individual gains indicates the magnitude of the gain to be applied to the corresponding individual input loudness level. The output loudness level after applying the individual gains represents the target loudness level for the audio content in the individual audio data frame in a particular playback environment.

In addition to specifying the mapping between gain and loudness levels, the Dynemic range compression curve may or provides specific release and attack times when applying specific gains. May be done. Attack refers to an increase in signal energy (or loudness) during successive time samples. Release, on the other hand, refers to a decrease in signal energy (or loudness) during successive time samples . Attack time (eg 10ms, 20ms, etc.) is the time constant used to smooth the DRC gain when the corresponding signal is in attack mode. Release time (eg 80 ms, 100 ms, etc.) is the time constant used to smooth the DRC gain when the corresponding signal is in release mode. In some embodiments, additional, optional or alternative, these time constants are used for smoothing the signal energy (loudness) before determining the DRC gain.

Different dynamic range compression curves can correspond to different playback environments. For example, the dynamic range compression curve for the playback environment of a flat panel TV may be different from the dynamic range compression curve for the playback environment of a portable device. In some embodiments, the reproduction device may have more than one reproduction environment. For example, the first dynamic range compression curve for the first playback environment of a portable device with speakers is the second dynamic range compression curve for the second playback environment of the same portable device with a headset. May be different.

<u style="single">&lt;3. Audio decoder&gt;</u> FIG. 1A shows an exemplary audio decoder 100 with a data extractor 104, a dynamic range controller 106, an audio renderer 108, and the like.

In some embodiments, the data extractor (104) is configured to receive the encoded input signal 102. The encoded input signal as described herein may be a bitstream containing encoded (eg, compressed) input audio data frames and metadata. The data extractor (104) is configured to extract / decode input audio data frames and metadata from the encoded input signal (102). Each input audio data frame has multiple encoded audio data blocks, each representing multiple audio samples. Each frame represents a (eg, constant) time interval that contains a certain number of audio samples. The frame size can vary with the sample rate and the encoded data rate. Audio samples are quantized audio data elements that represent spectral content in one, two or more (audio) frequency bands or frequency ranges (eg, input PCM samples, input time-frequency samples in the QMF matrix). Etc.). The quantized audio data element in the input audio data frame may represent the pressure wave in the digital (quantized) domain. The quantized audio data element can cover a finite range of loudness levels below the maximum possible value (eg clipping level, maximum loudness level, etc.).

The metadata can be used by a wide variety of receiving decoders for processing input audio data frames. Metadata is a variety of behavioral parameters related to one or more actions to be performed by the decoder (100), normalization parameters related to the dialog loudness level represented in the input audio data frame. Etc. may be included. Dialog loudness level is the loudness of a program (for example, a movie, a TV program, a radio broadcast, etc.), a part of a program, a dialog of a program, etc., such as dialog loudness, program loudness, average dialog loudness, etc. It can refer to any psychoacoustic or perceptual level.

Some or all of the behavior and functionality of the decoder (100) or its modules (eg, data extractor 104, dynamic range controller 106, etc.) responds to metadata extracted from the encoded input signal (102). May be adapted. For example, metadataincluding, but not limited to, dynamic range compression curves, dialog loudness levels, etc.is output by the decoder (100) in the digital domain for output audio data elements (eg output PCM samples, output time in the QMF matrix). -May be used to generate frequency samples, etc.). The output data element can then be used to drive an audio channel or speaker to achieve a specified loudness or reference playback level during playback in a particular playback environment.

In some embodiments, the dynamic range controller (106) receives some or all of the audio data elements and metadata in the input audio data frame and at least partially encoded the audio signal. Based on the metadata extracted from (102), the audio processing operation (for example, dynamic range control operation, gain smoothing operation, gain limiting operation, etc.) is performed on the audio data element in the input audio data frame. It is configured to execute and so on.

In some embodiments, the dynamic range controller (106) may include a selector 110, a loudness calculator 112, a DRC gain unit 114, and the like. The selector (110) is a speaker configuration (eg, flat panel mode, portable device with speakers, portable device with headphones, 5.1 speaker configuration, 7.1 speaker configuration) related to the specific playback environment in the decoder (100). Etc.), and a specific dynamic range compression curve may be selected from various dynamic range compression curves extracted from the encoded input signal (102).

The loudness calculator (112) may be configured to calculate the loudness level of one or more types represented by the audio data elements in the input audio data frame. Illustrative types of loudness levels are not limited to this, but are limited to individual loudness levels across individual frequency bands in individual channels over individual time intervals, and over a wide (or wide) frequency range in individual channels. Broadband (or broadband) loudness level, loudness level determined from an audio data block or frame or smoothed over an audio data block or frame, two or more audio data blocks or frames Includes any loudness level determined from or smoothed over two or more audio data blocks or frames, such as a loudness level smoothed over one or more time intervals. Zero, one or more of these loudness levels may be modified by the decoder (100) for dynamic range control.

To determine the loudness level, the loudness calculator (112) is one or more such as the spatial pressure level at a particular audio frequency represented by the audio data elements in the input audio data frame. Time-dependent physical loudness attributes can be determined. The loudness calculator (112) describes one or more types of loudness levels to derive one or more types of loudness levels based on one or more psychoacoustic functions that model human loudness perception. You can use the time-varying physical wave attributes of. Psychoacoustics functions are non-linear functions that transform a particular spatial pressure level at a particular audio frequency into a particular loudness for that particular audio frequency-built on a model of the human auditory system. And so on.

Loudness levels across multiple (audio frequencies) or multiple frequency bands (eg, broadband, broadband, etc.) may be derived through the integration of specific loudness levels over multiple (audio) frequencies or multiple frequency bands. .. Time averaging, smoothed loudness levels, etc. over one or more time intervals (eg, longer than represented by audio data elements in an audio data block or frame) are in the decoder (100). It may be obtained using one or more smoothing filters implemented as part of the audio processing operation.

In certain exemplary embodiments, specific loudness levels for different frequency bands may be calculated for each audio data block of a sample (eg, 256, for example). Prefix filters are used to apply frequency weighting (eg, similar to IEC B weighting) to specific loudness levels in integrating specific loudness levels into broadband loudness levels. May be used. The sum of wide loudness levels across two or more channels (eg, front left, front right, center, left surround, right surround, etc.) to provide the overall loudness level for the two or more channels. It may be executed.

In some embodiments, the overall loudness level may refer to a broadband loudness level in a channel (eg, central) in a speaker configuration. In some embodiments, the overall loudness level may refer to broadband (or broadband) loudness levels in multiple channels. The plurality of channels may be all channels in a certain speaker configuration. Additional, optional or alternative, the plurality of channels are a subset of channels in a speaker configuration (eg, a subset of channels including left front, right front and low frequency effects (LFE), left surround and right. It may include a subset of channels that include surround, a subset of channels that include the center, and so on.

The loudness level (eg broadband, broadband, global, specific) is for finding the corresponding DRC gain (eg static, pre-smoothing, pre-limit, etc.) from the selected dynamic range compression curve. It may be used as an input. The loudness level used as the input to find the DRC gain may first be adjusted or normalized with respect to the dialog loudness level from the metadata extracted from the encoded audio signal (102). In some embodiments, the adjustments and normalizations associated with adjusting the dialog loudness level are, but are not limited to, in the non-loudness region (eg, SPL region), in the encoded audio signal (102). In the encoded audio signal (102) for a portion of the audio content of the particular spatial pressure level represented in said portion of the audio content in the encoded audio signal (102). It may be performed before it is converted or mapped to the particular loudness level of the part of the audio content of.

In some embodiments, the DRC gain unit (114) is configured with a DRC algorithm to generate a gain (for example, for dynamic range control, for gain limiting, for gain smoothing, etc.) and to generate that gain. Applies to one or more loudness levels in one or more types of loudness levels represented by the audio data elements in the input audio data frame, and the target loudness for that particular playback environment. You may achieve the level. The application of gains as described in this paper (eg DRC gains) is not required, but may occur in the loudness region. In some embodiments, the gain is generated and smoothed based on a loudness calculation, which may be represented by a thorn or simply an SPL compensated for the dialog loudness level without conversion, for example. It may be applied directly to the input signal. In some embodiments, techniques such as those described herein apply gain to a signal in the loudness region, then convert the signal from the loudness region to the original (linear) SPL region, and in the loudness region the signal. The corresponding gain to be applied to the signal may be calculated by evaluating the signal before and after the gain is applied to the signal. The ratio (or difference when expressed in logarithmic dB) then determines the corresponding gain for that signal.

In some embodiments, the DRC algorithm works with multiple DRC parameters. The DRC parameters have already been calculated by an upstream encoder (eg 150) and embedded in the encoded audio signal (102) and by the decoder (100) in the encoded audio signal (102). Includes the dialog loudness level that can be obtained from the metadata. The dialog loudness level from the upstream encoder indicates the average dialog loudness level (for example, per program, for full-scale 1kHz sine wave energy, for reference square wave energy, etc.). In some embodiments, the dialog loudness level extracted from the encoded audio signal (102) may be used to reduce loudness level differences between programs. In certain embodiments, the reference dialog loudness level may be set to the same value between different programs in the same particular playback environment in the decoder (100). Output dialog loudness level averaged across multiple audio data blocks of a program is a reference dialog for the program (eg, preconfigured, system default, user configurable, profile dependent, etc.) Based on the dialog loudness level from the metadata, the DRC gain unit (114) transfers the dialog loudness related gain to each audio data block in the program so that it can be raised / lowered to the loudness level. Can be applied.

In some embodiments, the DRC gain addresses differences in loudness levels within the program by boosting or cutting signal parts in soft and / or loud sounds according to a selected dynamic range compression curve. May be used for. One or more of these DRC gains can be a selected dynamic range compression curve and (broadband, broadband, overall,) determined from one or more of the corresponding audio data blocks, audio data frames, etc. It may be calculated / determined by the DRC algorithm based on any specific) loudness level.

The loudness level used to determine the DRC gain (eg, static, before smoothing, before gain limiting, etc.) by searching the selected dynamic range compression curve is a short interval (eg, about). May be calculated on 5.3 milliseconds, etc.). The integration time of the human auditory system (for example, about 200 milliseconds) can be much longer. The DRC gain obtained from the selected dynamic range compression curve may be smoothed with a time constant to take into account the long integration time of the human auditory system. In order to achieve a fast rate of change (increase or decrease) in loudness level, a short time constant may be used to cause a change in loudness level in a short time interval corresponding to the short time constant. Conversely, in order to implement a slow rate of change (increase or decrease) in loudness level, a long time constant may be used to change the loudness level over a long time interval corresponding to the long time constant. ..

The human auditory system may respond to increasing loudness levels and decreasing loudness levels with different integration times. In some embodiments, different times depending on whether the loudness level is increasing or decreasing in order to smooth the static DRC gains retrieved from the selected dynamic range compression curve. Constants may be used. For example, according to the characteristics of the human auditory system, attack (increased loudness level) is smoothed with a relatively short time constant (eg attack time), while release (decreased loudness level) is. Smoothed with a relatively long time constant (eg release time).

The DRC gain for a portion of the audio content (eg, one or more audio data blocks, audio data frames, etc.) uses the loudness level determined from said portion of the audio content. It may be calculated. The loudness level to be used for searching on the selected dynamic range compression curve is first in the metadata extracted from the encoded audio signal (102) (eg, a program of which that audio content is part). It may be adjusted with respect to the dialog loudness level (for example, in relation to it).

Reference dialog loudness level (for example, -31 dB in "line" mode_FS-20dB in "RF" mode_FSEtc.) may be specified or established for a particular playback environment in the decoder (100). Additional, alternative or optionally, in some embodiments, the user may be given control over setting or changing the reference dialog loudness level in the decoder (100).

The DRC gain unit (114) now determines the dialog loudness-related gain for audio content to cause a change from the input dialog loudness level to the reference dialog loudness level as the output dialog loudness level. Can be configured.

In some embodiments, the DRC gain unit (114) may be configured to handle peak levels in a particular playback environment in the decoder (100), adjusting the DRC gain to prevent clipping. In some embodiments, under the first approach, the audio content extracted from the encoded audio signal (102) is a reference multi-channel with more channels than the channels of a particular speaker configuration in the decoder. If the audio data elements for the configuration are included, a particular speaker configuration downmix may be performed from the reference multichannel configuration before determining and processing the peak level to prevent clipping. Additional, optional or alternative, in some embodiments, under the second approach, the audio content extracted from the encoded audio signal (102) is of a particular speaker configuration in the decoder. If it contains audio data elements for a reference multi-channel configuration with more channels than channels, then the downmix equation (eg, for example) to obtain the peak level for a particular speaker configuration in the decoder (100). , ITU stereo downmix, matrixed-surround compatible downmix downmix) etc.) may be used. The peak level may be adjusted to reflect the change from the input dialog loudness level to the reference dialog loudness level as the output dialog loudness level. The maximum permissible gain that does not cause clipping (for example, for an audio data block, for an audio data frame, etc.) is at least partially the reciprocal of the peak level (for example, multiplied by -1). It may be determined based on (such as being done). Thus, the audio decoder based on the techniques described herein can be configured to accurately determine the peak level and apply anti-clipping specifically for the playback configuration on the decoder side. Neither the audio decoder nor the audio encoder needs to make assumptions about the worst-case scenario in the hypothetical decoder. In particular, the decoder in the first approach above does not use the downmix equation, downmix channel gain, etc. (which is used under the second approach as described above) and peaks exactly. You can determine the level and apply anti-clipping after downmixing.

In some implementations, the combination of dialog loudness level and adjustment to DRC gain is the worst-case downmix (for example, one that produces the maximum peak level after downmix, maximum downmix channel gain). Prevent peak level clipping, even in those that generate. However, in some other embodiments, the combination of dialog loudness level and adjustment to DRC gain may not be sufficient to prevent peak level clipping. In these embodiments, the DRC gain may be replaced by the highest gain that prevents clipping at the peak level (eg, capped).

In some embodiments, the DRC gain unit (114) is configured to obtain a time constant (eg, attack time, release time, etc.) from metadata extracted from the encoded audio signal (102). The DRC gain, time constant, maximum allowable gain, etc. may be used by the DRC gain unit (114) to perform DRC, gain smoothing, gain limiting, etc.

For example, the application of DRC gain may be smoothed with a filter controlled by a time constant. The gain limiting operation may be implemented by the min () function, which takes the smaller of the gain to be applied and the maximum permissible gain for that gain. Through this function, the gain (eg, pre-limit, DRC, etc.) may be immediately replaced by the maximum permissible gain, such as over a relatively short time interval. This prevents clipping.

In some embodiments, the audio renderer (108) applies gains determined based on DRC, gain limiting, gain smoothing, etc. to the input audio data extracted from the encoded audio signal (102). It is then configured to generate channel-specific audio data (116) for that particular speaker configuration (eg, multi-channel). The channel-specific audio data (118) may be used to drive the speakers, headphones, etc. represented in the speaker configuration.

Additional and / or optionally, in some embodiments, the decoder (100) performs one or more other actions related to pre-processing, post-processing, rendering, etc. related to the input audio data. Can be configured to run.

The techniques described in this article include a variety of speaker configurations that support a wide variety of different surround sound configurations (eg, 2.0, 3.0, 4.0, 4.1, 4.1, 5.1, 6.1, 7.1, 7.2, 10.2, 10-60 speaker configurations, Used with 60+ speaker configurations, object signals or combinations of object signals, etc.) and a variety of different rendering environment configurations (eg cinemas, parks, opera houses, concert halls, bars, homes, auditoriums, etc.) Can be done.

<u style="single">&lt;4. Audio encoder&gt;</u> FIG. 1B shows an exemplary encoder 150. The encoder (150) may include an audio content interface 152, a dialog loudness analyzer 154, a DRC reference storage 156, an audio signal encoder 158, and the like. The encoder 150 may be part of a broadcast system, an internet-based content server, an over-the-air network operator system, a movie production system, and the like.

In some embodiments, the audio content interface (152) receives audio content 160, audio content control input 162, etc., and at least in part audio content (160), audio content control input. It is configured to generate an encoded audio signal (eg 102) based on part or all of (162). For example, the audio content interface (152) may be used to receive audio content (160), audio content control input (162) from content creators, content providers, and so on.

The audio content may be part or all of the overall media data, including audio only, audiovisual, and the like. The audio content (160) may include one or more of parts of a program, a program, some programs, one or more commercials, and the like.

In some embodiments, the dialog loudness analyzer (154) is one or more parts of the audio content (152) (eg, one or more programs, one or more commercials, etc.). Or it is configured to determine / establish multiple dialog loudness levels. In some embodiments, the audio content is represented by one or more sets of audio tracks. In some embodiments, the audio content dialog audio content is on a separate audio track. In some embodiments, at least a portion of the audio content is on an audio track that includes non-dialog audio content.

The audio content control input (162) is part of a user control input, a control input provided by an external system / device to the encoder (150), a control input from the content creator, a control input from the content provider, etc. Or it may include all. A user, such as a mixing engineer, can provide / specify one or more dynamic range compression curve identifiers. These identifiers may be used to extract one or more dynamic range compression curves that best fit the audio content (160) from a data store such as the DRC reference store (156).

In some embodiments, the DRC reference reservoir (156) is configured to store the DRC reference parameter set and the like. The set of DRC reference parameters may contain definition data for one or more dynamic range compression curves and the like. In some embodiments, the encoder (150) may encode two or more dynamic range compression curves (eg, in parallel) into an encoded audio signal (102). Zero, one or more of those dynamic range compression curves may be standard-based, proprietary, customized, decoder-modifiable, and so on. In one exemplary embodiment, both the dynamic range compression curves of FIGS. 2A and 2B can be encoded (eg, in parallel) into the encoded audio signal (102).

In some embodiments, the audio signal encoder (158) receives audio content from the audio content interface (152), dialog loudness level from the dialog loudness analyzer (154), and so on, and the DRC. Extract one or more DRC reference parameter sets from the reference reservoir (156), format the audio content into audio data blocks / frames, and metadata the dialog loudness level, DRC reference parameter set, etc. , Metadata container, metadata field, metadata structure, etc.) and configured to encode audio data blocks / frames and metadata into an encoded audio signal (102), etc. Can be done.

The audio content to be encoded in the audio signal encoded as described in this article is one of a variety of methods, such as wirelessly, over a wired connection, through a file, via an internet download, etc. Can be received in one or more of multiple and diverse source audio formats.

The encoded audio signals described in this article may be part of the overall media data bitstream (for example, for audio broadcasts, audio programs, audiovisual programs, audiovisual broadcasts, etc.). it can. Media data bitstreams can be accessed from servers, computers, media storage devices, media databases, media files, and more. Media data bitstreams may be broadcast, transmitted or received over one or more wireless or wired network links. Media data bitstreams are delivered through one or more intermediaries such as network connections, USB connections, wide area networks, local area networks, wireless connections, optical connections, buses, crossbar connections, serial connections, etc. It may be communicated.

Any of the components depicted (eg, Figure 1A, Figure 1B, etc.) may be hardware, software, or a combination of hardware and software, one or more processes and / or one or more IC circuits. It may be implemented as (eg ASIC, FPGA, etc.).

<u style="single">&lt;5. Dynamic range compression curve&gt;</u> Figures 2A and 2B show exemplary dynamic range compression curves that can be used by the DRC gain unit (104) in the decoder (100) to derive the DRC gain from the input loudness level. As shown, the dynamic range compression curve may be centered on the reference loudness level in the program to provide the appropriate overall gain for a particular playback environment. Illustrative definition data for dynamic range compression curves (eg, in the metadata of the encoded audio signal 102) (eg, but not limited to, including, but not limited to, boost ratio, cut ratio, attack time, release time, etc.) Is shown in the table below. Here, each profile in a plurality of profiles (for example, film standard, film light, music standard, music light, speech, etc.) is Represents a particular playback environment (for example, in Decoder 100).

<tables num="1"><img file="JP6633232B2_D0001.tif" /></tables> Some embodiments are dB_SPLOr dB_FSLoudness level and dB represented by_SPLYou may receive one or more compression curves described using the gain expressed in dB with respect to. On the other hand, the DRC gain is dB_SPLIt is performed with a different loudness representation (eg, thorn) that has a non-linear relationship with the loudness level. In doing so, the compression curve used in the DRC gain calculation may be transformed to be described using the different loudness representation (eg, thorn).

<u style="single">&lt;6. DRC gain, gain limitation and gain smoothing&gt;</u> Figure 3 shows an exemplary processing logic for the combined DRC and limit gain determination / calculation. The processing logic may be implemented by a decoder (100), an encoder (150), or the like. For illustration purposes only, a DRC gain unit (eg 114) in a decoder (eg 100) may be used to implement the processing logic.

The DRC gain for a portion of the audio content (eg, an audio data block, one or more audio data frames, etc.) uses the loudness level determined from that portion of the audio content. It may be calculated. The loudness level is first about (for example, in relation to) the dialog loudness level in the metadata extracted from the encoded audio signal (102) (for example, a program of which that audio content is part). Etc.) may be adjusted. In the example shown in FIG. 3, the difference between said partial loudness level and said dialog loudness level (dialnorm) of the audio content finds the DRC gain from the selected dynamic range compression curve. May be used as input for.

To prevent clipping of the output audio data elements in that particular playback environment, the DRC gain unit (114) is specific to a particular combination of playback environments in the encoded audio signal 102 and decoder 100, etc. It may be configured to handle peak levels in a particular regeneration scenario. The reproduction scenario may be one of a variety of possible reproduction scenarios (eg, multi-channel scenarios, downmix scenarios, etc.).

In some embodiments, individual peaks for individual parts of audio content at a particular time resolution (eg, audio data blocks, some audio data blocks, audio data frames, etc.). Levels may be provided as part of the metadata extracted from the encoded audio signal (102).

In some embodiments, the DRC gain unit (114) can be configured to determine the peak level in these scenarios and adjust the DRC gain if necessary. A parallel process may be used by the DRC gain unit (114) to determine the peak level of audio content during the DRC gain calculation. For example, audio content may be encoded for a reference multi-channel configuration with more channels than the channels of the particular speaker configuration used by the decoder (100). The audio content for the more channels in the reference multi-channel configuration is downmixed audio data (eg, ITU stereo downmix, matrixed-surround compatible). It may be converted to (downmix), etc.) to derive fewer channels for a particular speaker configuration in the decoder (100). In some embodiments, under the first approach, downmixing from a reference multi-channel configuration to a particular speaker configuration is performed before determining and processing peak levels to prevent clipping. May be good. Additional, optional or alternative, in some embodiments, under the second approach, the downmix channel gain associated with downmixing the audio content is that of that particular speaker configuration. May be used as part of the input to adjust, derive, calculate, etc. the peak level for. In one exemplary embodiment, the downmix channel gain is one or more downmixes used to perform a downmix operation from a reference multi-channel configuration to a particular speaker configuration in the playback environment at decoder (100). It may be derived based on, at least in part, the equation of.

For some media applications, the browse dialog loudness level (for example, -31dB in "line" mode)_FS-20dB in "RF" mode_FSEtc.) may be specified or envisioned for a particular playback environment in the decoder (100). In some embodiments, the user may be given control over setting or changing the reference dialog loudness level in the decoder (100).

Dialog loudness-related gains may be applied to the audio content to adjust the dialog loudness level (for example, in the output) to the referenced dialog loudness level. To reflect this adjustment, the peak level should be adjusted accordingly. In one example, the (input) dialog loudness level is -23dB_FSIt may be. Reference dialog loudness level is -31dB_FSIn "line" mode of, the adjustment to the (input) dialog loudness level is -8 dB to produce the output dialog loudness level of the reference dialog loudness level. In this "line" mode, the adjustment to the peak level is also -8dB, which is the same as the adjustment to the dialog loudness level. Reference dialog loudness level is -20dB_FSIn "RF" mode of, the adjustment to the (input) dialog loudness level is 3 dB to produce the output dialog loudness level of the reference dialog loudness level. In this "RF" mode, the adjustment to the peak level is also 3 dB, which is the same as the adjustment to the dialog loudness level.

The difference between the peak level and the reference dialog loudness level (denoted as "dialref") and the dialog loudness level ("dialnorm") in the metadata from the encoded audio signal (102). The sum of may be used as an input to calculate the maximum (eg, acceptable) gain for the DRC gain. The adjusted peak level is (0dB)_FS(For clipping level of) dB_FSThe maximum permissible gain without clipping (for example, for the current audio data block, for the current audio data frame, etc.) is simply the reciprocal of the adjusted peak level. There is (for example, multiplied by -1).

In some embodiments, the peak level is the clipping level (0 dB), even if the dynamic range compression curve from which the DRC gain was derived was designed to cut some loud sounds._FS(Represented by) may be exceeded. In some embodiments, the combination of dialog loudness level and adjustment to DRC gain is at peak level, even in the worst case downmix (for example, one that produces the maximum downmix channel gain). Prevent clipping. However, in some other embodiments, the combination of dialog loudness level and adjustment to DRC gain may not be sufficient to prevent peak level clipping. In these embodiments, the DRC gain may be replaced by the highest gain that prevents clipping at the peak level (eg, capped).

In some embodiments, the DRC gain unit (114) is configured to obtain a time constant (eg, attack time, release time, etc.) from metadata extracted from the encoded audio signal (102). These time constants may or may not change with one or more of the audio content's dialog loudness level or the current loudness level. The DRC gain, time constant, and maximum gain retrieved from the dynamic range compression curve may be used to perform gain smoothing and limiting operations.

In some embodiments, the DRC gain, which may potentially be gain limited, does not exceed the maximum peak loudness level in a particular reproduction environment. The static DRC gain derived from the loudness level may be smoothed with a time constant controlled filter. Restricted behavior may be implemented by one or more min () functions. Through this function, the DRC gain (eg, before the limit) may be immediately replaced by the maximum permissible gain, such as over a relatively short time interval. This prevents clipping. The DRC algorithm may be configured to smoothly release from said clipping gain to lower gain as the peak level of the incoming audio content shifts from above the clipping level to below the clipping level.

One or more different implementations (eg, real-time, two-pass, etc.) may be used to perform the determination / calculation / application of the DRC gain shown in Figure 3. For illustration purposes only, adjustments to dialog loudness levels, DRC gains (such as static), time-dependent gain fluctuations due to smoothing, gain clipping due to limitations, etc. can be made from the DRC algorithm above. Has been described as the combined gain of. However, in various embodiments, to prevent clipping due to dialog loudness level control (eg, between different programs), dynamic range control (eg, for different parts of the same program). Other approaches may be used to apply gain to audio content, such as for gain smoothing. For example, some or all of the adjustments to dialog loudness levels, DRC gains (such as static), time-dependent gain fluctuations due to smoothing, gain clipping due to limitations, etc. are partial / individual. It can be applied to, serially, parallelly, partially in series, partially in parallel, and so on.

<u style="single">&lt;7. Input smoothing and gain smoothing&gt;</u> In addition to DRC gain smoothing, in various embodiments, other smoothing processes under the techniques described herein may be implemented. In one example, input smoothing may be used, where the input audio data extracted from the encoded audio signal (102) is smoothed using, for example, a simple unipolar smoothing filter, without input smoothing. Specific loudness level spectra may be obtained with specific loudness level spectra that have better temporal characteristics (eg, smoother in time, less spikes in time, etc.).

In some embodiments, the different smoothing processes described herein can use different time constants (eg 1 second, 4 seconds, etc.). In some embodiments, two or more smoothing processes can use the same time constant. In some embodiments, the time constants used in the smoothing process described herein may be frequency dependent. In some embodiments, the time constants used in the smoothing process described herein may be frequency independent.

One or more smoothing processes may be connected to a reset process that supports automatic or manual resetting of the one or more smoothing processes. In some embodiments, when a reset occurs in the reset process, the smoothing process may accelerate the smoothing operation by switching to or shifting to a smaller time constant. In some embodiments, the memory of the smoothing process may be reset to a value when the reset occurs in the reset process. This value may be the last input sample to the smoothing process.

<u style="single">&lt;8. DRC over multiple frequency bands&gt;</u> In some embodiments, specific loudness levels in specific frequency bands can be used to derive the corresponding DRC gains in those specific frequency bands. However, this can lead to changes in timbre. These specific loudness levels can vary significantly in different bands, thus suffering different DRC gains, even when broadband (or broadband) loudness levels across the entire frequency band remain constant. Is.

In some embodiments, instead of applying a DRC gain that changes with the individual frequency band, a DRC gain that does not change with the frequency band but changes over time is applied instead. The same time-varying DRC gain is applied over the entire frequency band. The time-averaged DRC gain of time-varying DRC gain is from the selected dynamic range compression curve based on broadband, broadband and / or overall loudness levels over a broadband (or broadband) range or multiple frequency bands. It may be set to be the same as the derived static DRC gain. As a result, changes to the timbral effect that can be caused by applying different DRC gains in different frequency bands in other approaches can be prevented.

In some embodiments, the DRC gains in the individual frequency bands are controlled using broadband (or broadband) DRC gains that are determined based on the broadband (or broadband) loudness level. DRC gains in individual frequency bands may operate around broadband (or broadband) DRCs found in dynamic range compression curves based on broadband (or broadband) loudness levels. Thus, the DRC gain in individual frequency bands time-averaged over a time interval (for example, longer than 5.3 ms, 20 ms, 50 ms, 80 ms, 100 ms, etc.) is in the dynamic range compression curve. Same as the broadband level shown. In some embodiments, loudness level fluctuations over a short time interval with respect to the time interval, which deviate from the time averaged DRC gain, are acceptable between the channels and / or frequency bands. This approach ensures the application of the correct multi-channel and / or multi-band time-averaged DRC gains shown in the dynamic range compression curve, and such DRC gains in short time intervals are shown in the dynamic range compression curves. Prevents deviations from the time-averaged DRC gain too much.

<u style="single">&lt;9. Volume adjustment in the loudness area&gt;</u> Applying linear processing for volume regulation to audio excitation signals under other approaches that do not implement the techniques described in this article can make low audible signal levels inaudible (eg, humans). Below the frequency-dependent auditory threshold of the auditory system).

Under the techniques described in this article, volume adjustment of audio content is in the physical domain (eg dB)._SPLIt can be done or implemented in the loudness domain (with, for example, Thorne's representation) rather than (with representation, etc.). In some embodiments, the loudness level of all bands is loudness in order to maintain the perceptual quality and / or integrity of the loudness level relationship among all bands at all volume levels. Scaled with the same factors in the region. Volume adjustments based on setting and adjusting gains in the loudness region described in this paper are non-linear in the physical region (or in the digital region representing the physical region) that apply different scaling factors to audio excitation signals in different frequency bands. It may be converted back to processing and implemented through the non-linear processing. Non-linear processing in the physical domain converted from volume adjustment in the loudness domain under the techniques described in this paper has a DRC gain that prevents most or all of the low audible levels in the audio content from becoming inaudible. Attenuates or improves the loudness level of content. In some embodiments, the difference in loudness levels between loud and soft sounds in the program provides these DRC gains that keep low audible signal levels above the auditory threshold of the human auditory system. Used to reduce-but not perceptually eliminated. In some embodiments, frequencies or frequencies having an excitation signal level close to the auditory threshold at low volume levels to maintain spectral perception and similarities such as perceived timbre over a large range of volume levels. The band is less attenuated and is therefore perceptually audible.

The techniques described in this paper are conversions between signal levels, gains, etc. in the physical domain (or digital domain representing the physical domain) and loudness levels, gains, etc. in the loudness domain (eg, back and forth conversions, etc.). ) May be implemented. These transformations are based on forward and inverse versions of one or more nonlinear functions (eg, mappings, curves, piecewise linear segments, look-up tables, etc.) built on models of the human auditory system. You may.

<u style="single">&lt;10. Gain profile by differential gain&gt;</u> In some embodiments, the audio encoders described herein (such as 150) are configured to provide profile-related metadata to downstream audio decoders. For example, profile-related metadata may be carried in the encoded audio signal as part of the audio-related metadata along with the audio content.

The profile-related metadata described in this paper includes, but is not limited to, definition data for multiple gain profiles. The one or more first gain profiles in the plurality of gain profiles (denoted as one or more default gain profiles) are one or more corresponding DRC curves (one or more default DRC curves). Is represented by). The definition data is included in the profile-related metadata. The one or more second gain profiles in the plurality of gain profiles (denoted as one or more non-default gain profiles) are one or more of the differential gains for the one or more default DRC curves. Represented by the corresponding set of. The definition data is included in the profile-related metadata. More specifically, the default DRC curve (for example, in profile-related metadata) can be used to represent the default gain profile, and the differential gain for the default gain profile (for example, in profile-related metadata). The set can be used to represent the non-default gain profile.

In some embodiments, the set of differential gains representing the non-default gain profile in relation to the default DRC curve representing the default gain profile is the non-differential (eg, non-default, etc.) generated for the non-default gain profile. ) Includes a gain difference (or gain adjustment) between a set of gains and a set of non-differential (eg, default) gains generated for the default gain profile. Examples of non-differential gains include, but are not limited to, null gain, DRC gain or attenuation, gain or attenuation for dialog normalization, gain or attenuation for gain limitation, gain or attenuation for gain smoothing, and the like. The gains described herein (eg, non-differential gains, differential gains, etc.) may be time-dependent or may have values that change over time.

To generate a set of non-differential gains for a gain profile (eg, default gain profile, non-default gain profile, etc.), the audio encoders described in this article perform a set of gain generation operations specific to the gain profile. You may do it. The set of gain generation operations may include a DRC operation, a gain limiting operation, a gain smoothing operation, and the like. It is (1) globally applicable to all gain profiles; (2) unique to one or more but not all gain profiles, unique to one or more default DRC curves; (3) one or more. Unique to non-default DRC curves; (4) Specific to corresponding (eg default, non-default, etc.) gain profiles; (5) Parameterization limits supported by media encoding formats, media standards, media-specific specifications, etc. It relates to one or more of the algorithms, curves, functions, behaviors, parameters, etc. that exceed; (6) One of the algorithms, curves, functions, behaviors, parameters, etc. Including, but not limited to, any of the actions relating to one or more.

In some embodiments, the audio decoder (150), at least in part, has a default gain profile represented by a default DRC curve (eg, by definition data in the profile-related metadata of the encoded audio signal) and said. A set of differential gains for audio content (152) is determined based on a non-default gain profile that is different from the default gain profile, and the set of differential gains is the non-default gain profile (eg, the default DRC curve). As a representation (for, etc.), it can be configured to be included as part of profile-related metadata in the encoded audio signal. The set of differential gains extracted from the profile-related metadata in the encoded audio signal in relation to the default DRC curve is by the receiving audio decoder and by the set of differential gains in relation to the default DRC curve. It can be used to efficiently and consistently perform gain operation (or attenuation operation) in a reproduction environment or scenario for the particular gain profile represented. This allows the receiving audio decoder to apply gain or attenuation for that particular gain profile without requiring the receiving audio decoder to implement a set of gain-generating operations. To generate the gain or attenuation, a set of gain generation operations can be implemented in the audio encoder (150).

In some embodiments, one or more sets of differential gains may be included in the profile-related metadata by the audio encoder (150). Each of one or more sets of differential gains has a corresponding non-default gain profile in one or more non-default gain profiles in relation to the corresponding default gain profile in one or more default gain profiles. It may be derived from. For example, the first set of differential gains in said one or more sets of differential gains may be derived from the first non-default gain profile in relation to the first default gain profile, while the differential. A second set of differential gains in those sets of gains may be derived from a second non-default gain profile in relation to the second default gain profile.

In some embodiments, the first set of differential gains is the first set of non-differential non-default gains generated based on the first non-default gain profile and the first default gain profile. Includes the first gain difference (or gain adjustment) determined to and from the first set of non-differential default gains generated based on. On the other hand, the second set of differential gains is generated based on the second set of non-default non-default gains generated based on the second non-default gain profile and the second default gain profile. Includes a second gain difference determined with and from a second set of non-differential default gains.

The first default gain profile and the second default gain profile may be the same (eg, represented by the same default DRC curve with the same set of gain generation operations) or different (for example). For example, represented by a different default DRC curve, represented by a default DRC with different sets of gain-generating behaviors, and so on). In various embodiments, additional, optional or alternative, the first non-default gain profile may or may not be the same as the second non-default gain profile.

The profile relationship metadata generated by the audio encoder (150) is one or more to indicate the existence of one or more sets of differential gains for one or more corresponding non-default gain profiles. It can carry specific flags, indicators, data fields, and so on. Profile-related data may also include preference flags, indicators, data fields, etc. to indicate which non-default gain profile is preferred to render the audio content in a particular playback environment or scenario. Good.

In some embodiments, the audio decoder described herein (such as 100) is configured to decode audio content (such as multi-channel) from the encoded audio signal (102). Extract the dialog loudness level (for example, "dial norm") from the loudness metadata delivered with the audio content.

In some embodiments, the audio decoder (for example, 100) performs at least a set of gain generation operations on a gain profile such as the first default profile, the second default profile, and so on. It is composed. For example, the audio decoder (100) decodes an encoded audio signal (102) with a dialog loudness level (for example, "dialnorm"); it is encoded by performing a set of gain generation operations. Obtains a set of non-differential default gains (or attenuations) for the default gain profile represented by the default DRC curve from which the definition data can be extracted from the audio signal (102) by the audio decoder (100); during decoding Apply the set of non-differential default gains for the default gain profile (for example, the difference between the reference loudness level and the "dialnorm") to the output dialog loudness level of the sound output to refer to the loudness level. Align / adjust; etc.

Additional, optional or alternative, in some embodiments, the audio decoder (100) is configured to extract at least one set of differential gains from the encoded audio signal (102). The set of differential gains represents a non-default gain profile in relation to the default DRC curve as discussed above as part of the metadata delivered with the audio content. In some embodiments, the profile relationship metadata comprises one or more different sets of differential gains, each of which is a default DRC representing a default gain profile. Represents a non-default gain profile in relation to the curve. The presence of a DRC curve or set of differential gains in the profile-related metadata may be indicated by one or more flags, indicators, and data fields carried in the profile-related metadata.

In response to determining that the one or more sets of differential gains are present, the audio decoder (100) is out of the one or more different sets of differential gains of a particular non-default. The set of differential gains corresponding to the gain profile can be determined / selected. The audio decoder (100) furtherfor example, among the definition data for one or more different default DRC curves in profile-related metadatais the reference by which the set of differential gains represents the particular gain profile. It can be configured to identify the default DRC curve.

In some embodiments, the audio decoder (100) is configured to perform a set of gain generation operations to obtain a set of non-differential default gains (or attenuations) for said default gain profile. The set of gain-generating operations performed by the audio decoder (100) to obtain the set of non-differential default gains based on the default DRC curve may relate to one or more, standard, proprietary, etc. It may include one or more actions. In some embodiments, the audio decoder (100) is generated by the set of differential gains from which definition data is extracted from profile-related metadata and the set of gain generation operations based on the default DRC curve. Generates a set of non-differential non-default gains for the particular non-default gain profile based on said set of non-differential default gains; during decoding, non-differential non-default gains for that default gain profile (eg, for example. The set of reference loudness levels (such as the difference between the "dialnorm") is applied to align / adjust the output dialog loudness level of the sound output to the reference loudness level; etc.

In some embodiments, the audio decoder (100) can perform gain-related operations on one or more gain profiles. The audio decoder (100) can be configured to determine and perform gain-related actions for a particular gain profile based on one or more factors. These factors are not limited to: user inputs that specify preferences for a particular user-selected gain profile, user inputs that specify preferences for system-selected gain profiles, audio decoders (100). Features of a particular speaker or audio channel configuration used by, features of an audio decoder (100), availability of profile-related metadata for said particular gain profile, any encoder-generated preference for a gain profile. It may contain one or more such as flags. In some embodiments, the audio decoder (100) implements one or more procedural rules to determine or select a particular gain profile when there is a conflict between these factors. Alternatively, further user input may be requested.

<u style="single">&lt;11. Additional operation related to gain&gt;</u> Under the techniques described in this article, other processes such as dynamic equalization, noise compensation, etc. should also be performed in the loudness (eg, perceptual) domain rather than in the physical domain (or the digital domain that represents the physical domain). Can be done.

In some embodiments, gains from some or all of various processes such as DRC, equalization noise compensation, anti-clip, gain smoothing, etc. may be combined with the same gain in the loudness region and / or It may be applied in parallel. In some other embodiments, the gain from some or all of the various processes such as DRC, equalization noise compensation, anti-clip, gain smoothing, etc. may be separate gains in the loudness region, and / Or at least in part may be applied in series. In some other embodiments, gains from some or all of the various processes such as DRC, equalization noise compensation, anti-clip, gain smoothing, etc. may be applied in sequence.

<u style="single">&lt;12. Specific and broadband (or broadband) loudness levels&gt;</u> One or more audio processing elements, units, components, etc. such as transmission filters, auditory filter banks, synthetic filter banks, short-time Fourier transforms, etc., by encoders or decoders to perform the audio processing operations described in this article. May be used.

In some embodiments, one or more transmission filters that model the filtering of the outer and middle ear of the human auditory system come in an audio signal (eg, encoded audio signal 102, from a content provider). May be used to filter audio content, etc.). In some embodiments, an auditory filter bank may be used to model the frequency selectivity and frequency spread of the human auditory system. Excitation signal levels from some or all of these filters are determined / calculated and smoothed with a frequency-dependent time constant that becomes shorter over higher frequencies to model the integral of energy in the human auditory system. May be done. A non-linear function (eg, relationship, curve, etc.) between the excitation signal and the specific loudness level may then be used to obtain a profile of the frequency-dependent specific loudness level. Broadband (or broadband) loudness levels can be obtained by integrating specific loudness across frequency bands.

A straightforward sum / integral of a particular loudness level (for example, using weights equal to all frequency bands) can work well for broadband signals. However, such an approach may underestimate the loudness level (eg, perceptual) for a narrowband signal. In some embodiments, specific loudness levels at different frequencies or in different frequency bands are given different weights.

In some embodiments, the auditory filter bank and / or transmission filter as described above may be replaced by one or more short-time Fourier transforms (STFTs). The response of the transfer filter and the auditory filter bank may be applied in the Fast Fourier Transform (FFT) region. In some embodiments, for example, when one or more transmission filters (eg, forward) are used in or prior to the conversion from the physical domain (or the digital domain representing the physical domain) to the loudness domain. One or more reverse transmission filters are used. In some embodiments, the reverse transmission filter is not used, for example, when STFT is used in place of the auditory filter bank and / or transmission filter. In some embodiments, the auditory filter bank is omitted; instead, one or more quadrature mirror filters (QMF) are used. In these embodiments, the diffusion effect of the basement membrane in the model of the human auditory system can be omitted without significantly affecting the matters of audio processing behavior described herein.

Under the techniques described herein, different numbers of frequency bands (eg, 20 frequency bands, 40 frequency bands, etc.) may be used in different embodiments. Additional, optional or alternative, different bandwidths may be used in different embodiments.

<u style="single">&lt;13. Individual gains for individual subsets of channels&gt;</u> In some embodiments, when a particular speaker configuration is a multi-channel configuration, the excitation signals of all channels are first added before the conversion from the physical region (or the digital region representing the physical region) to the loudness region. May provide an overall loudness level. However, applying the same gain to all channels in a particular speaker configuration can result in a spatial balance between different channels in that particular speaker configuration (such as the relative loudness level between different channels). May not be saved.

In some embodiments, each loudness level and its respective loudness level are preserved so that the relative perceptual loudness levels between different channels can be optimally or correctly maintained. The corresponding gain obtained based on may be determined or calculated for each channel. In some embodiments, the corresponding gains obtained based on the respective loudness levels are not equal to the same overall gain. For example, each of some or all of the corresponding gains may be equal to the overall gain plus a small correction (eg, channel-specific).

In some embodiments, to preserve spatial balance, each loudness level and the corresponding gain obtained based on that respective loudness level may be determined or calculated for each subset of the channel. .. In some embodiments, the corresponding gains obtained based on the respective loudness levels are not equal to the same overall gain. For example, each of some or all of the corresponding gains may be equal to the overall gain plus a small correction (eg, channel-specific). In some embodiments, a subset of channels may include two or more channels that are a true subset of all channels in that particular speaker configuration (eg, left front, right front and low frequency effects). A subset of channels containing (LFE), a subset of channels containing left and right surround, etc.). The audio content for a subset of channels may be a submix of the overall mixture carried in the encoded audio signal (102). The channels in the submix can be applied with the same gain.

In some embodiments, the signal level in the digital domain is the corresponding physics in the physical domain represented by the digital domain in order to generate the actual loudness (eg, actually perceived) from a particular speaker configuration. (For example, dB_SPLOne or more calibration parameters may be used to relate to the spatial pressure) level, such as. The one or more calibration parameters may be given values specific to the physical sound equipment in a particular speaker configuration.

<u style="single">&lt;14. Auditory scene analysis&gt;</u> In some embodiments, the encoders described herein detect an auditory event boundary in audio content (eg, encoded in an encoded audio signal 102) and have one or more ASA parameters. To generate and format the one or more ASA parameters as part of an encoded audio signal (eg 102) delivered to a downstream device (eg Decoder 100). Auditory scene analysis (ASA) may be implemented. ASA parameters can include, but are not limited to, the location of the auditory event boundary, the value of the auditory event certainty index (discussed further below), and the like.

In some implementations, the (eg, temporal) position of the auditory event boundary may be indicated in the encoded metadata within the encoded audio signal (102). Additional, optional or alternative, the position of the auditory event boundary (eg, temporal) is in the audio data block and / or frame where the position of the auditory event boundary is detected (eg, flag, data field). May be instructed (using, etc.).

In the usage herein, the auditory event boundary refers to the point where the preceding auditory event ends and / or the subsequent auditory event begins. Each auditory event occurs between two successive auditory event boundaries.

In some embodiments, the encoder (150) is configured to detect the auditory event boundary by the difference in a specific loudness spectrum between two consecutive (eg, temporal) audio data frames. To. Each specific loudness spectrum may include an unsmoothed loudness spectrum calculated from the corresponding audio data frame of those successive audio data frames.

In some embodiments, the specific loudness spectrum N [b, t] is a normalized specific loudness spectrum N, as shown in_NORMIt may be standardized to obtain [b, t].

N_NORM[b, t] = N [b, t] / max_b{N [b, t]} (1) where b indicates bandwidth, t indicates time or audio frame index, max_b{N [b, t]} is the maximum specific loudness level over all frequency bands.

The normalized specific loudness spectra are subtracted from each other and used to derive the difference absolute sum D [t], as shown in the following equation.

D [t] = Σ_b| N_NORM[b, t] -N_NORM[b, t-1] | (2) The absolute difference sum is mapped to an auditory event certainty index with a value range of 0 to 1 as follows.

<maths num="1"><img file="JP6633232B2_D0002.tif" /></maths>Where D_minAnd D_maxAre the minimum and maximum thresholds (for example, user-configurable, system-configurable, set in relation to the past value distribution of D [t] in the audio content, etc.).

In some embodiments, the encoder (150) has a D [t] of D (eg, at a particular t)._minIs configured to detect the auditory event boundary (eg, at that particular t) above.

In some embodiments, the decoder described herein (such as 100) extracts an ASA parameter from an encoded audio signal (such as 102) and uses that ASA parameter as the perceptual distortion of the auditory event. Used to prevent unintended boosts of soft sounds and / or unintended cuts of loud sounds that cause.

The decoder (100) is configured to reduce or prevent unintended distortion of the auditory event by ensuring that the gain is closer to constant within the auditory event and constraining much of the gain change to near the auditory event boundary. May be done. For example, the decoder (100) responds to gain changes in an attack at the auditory event boundary (eg, loudness level increase) with a relatively small time constant (eg, equal to or shorter than the minimum duration of various auditory events). It may be configured to use things, etc.). Therefore, the gain change in the attack can be implemented relatively quickly by the decoder (100). On the other hand, the decoder (100) may be configured to use a relatively long time constant relative to the duration of the auditory event in response to gain changes at the release in the auditory event (eg, loudness level reduction). .. Thus, the gain change in release can be implemented relatively slowly by the decoder (100) so that the sound that should be perceived constant or gradually attenuated is not audibly or perceptually disturbed. is there. A rapid response in attack at the auditory event boundary and a slow response in release at the auditory event allow fast perception of the arrival of the auditory event, as well as auditory events such as piano chords-this is a particular loudness level. Preserves perceptual quality and / or integrity between relationships and / or including loud and soft sounds linked by a particular temporal relationship.

In some embodiments, the auditory event and the auditory event boundary indicated by the ASA parameter control gain changes in one, two, part, or all of the channels in a particular speaker configuration in the decoder (100). Used by the decoder (100).

<u style="single">&lt;15. Loudness level transition&gt;</u> Loudness level transitions can occur, for example, between two programs, between programs and loud commercials. In some embodiments, the decoder (100) maintains a histogram of instantaneous loudness levels based on past audio content (for example, received from an audio signal 102 encoded over the last 4 seconds). It is composed. Two regions with increased probabilities can be recorded in the histogram over the time interval from before the loudness level transition to after the loudness level transition. One of those areas is centered on the previous loudness level, while the other of those areas is centered on the new loudness level.

The decoder (100) dynamically determines the smoothed loudness level as the audio content is processed, and the corresponding bin (eg, smoothed) of the histogram based on the smoothed loudness level. An instantaneous loudness level bin) containing the same value as the converted loudness level may be determined. The decoder (100) is further configured to compare the probabilities in the corresponding bin with thresholds (eg, 6%, 7%, 7.5%, etc.). Here, the total area of the histogram curve (for example, the sum of all bins) represents a 100% probability. The decoder can be configured to detect the occurrence of loudness level transitions by determining the probability that the corresponding bin will fall below the threshold. In response, the decoder (100) may be configured to select a relatively small time constant to adapt to the new loudness level relatively quickly. As a result, the duration of a loud (or soft) start within a loudness level transition can be reduced.

In some embodiments, the decoder (100) uses a silence / noise gate to prevent low instantaneous loudness levels from entering the histogram and becoming a high probability bin in the histogram. Additional, optional or alternative, the decoder (100) may be configured to use said ASA parameters to detect auditory events that should be included in the histogram. In some embodiments, the decoder (100) has a time-dependent value of the time-averaged auditory event certainty index.<maths num="2"><img file="JP6633232B2_D0003.tif" /></maths>May be determined from the ASA parameters. In some embodiments, the decoder (100) has a time-averaged auditory event certainty, such as a time-dependent value A [t] of an auditory event certainty index (such as instantaneous) from ASA parameters. Indicator value<maths num="3"><img file="JP6633232B2_D0004.tif" /></maths>May be calculated. The decoder (100) is the time-averaged auditory event certainty index at the same time as the loudness level.<maths num="4"><img file="JP6633232B2_D0005.tif" /></maths>Is below the histogram inclusion threshold (eg 0.1, 0.12, etc.) and may be configured to exclude the loudness level from entering the histogram.

In some embodiments, for loudness levels (eg, instantaneous) that are allowed to be included in the histogram (for example, the value of A [t] with the corresponding CLK is above the histogram inclusion threshold). , The loudness level is the same as, or proportional to, the time-dependent value of the time-averaged auditory event certainty index [A [t] with a histogram] at the same time as those loudness levels. Assigned weights. As a result, loudness levels near the auditory event boundary have more effect on the histogram than other loudness levels not near the auditory event boundary (for example, A [t] with CLK is relatively). Has a large value, etc.).

<u style="single">&lt;16. Reset&gt;</u> In some embodiments, the encoders described herein (such as 150) are configured to detect a reset event and include an indicator of the reset event in the encoded audio signal (eg 102). To. In the first example, the encoder (150) resets in response to determining that a continuous period of relative silence (eg, 250 ms configurable by the system and / or user) occurs. Detect events. In the second example, the encoder (150) detects a reset event in response to determining that a large momentary drop in excitation level occurs across all frequency bands. In the third example, the encoder is given input (eg user input, system controlled metadata, etc.) where a content transition (eg program start / end, scene change, etc.) occurs that requires a reset.

In some embodiments, the decoders described herein (such as 100) implement a reset mechanism that can be used to speed up gain smoothing in an instant. The reset mechanism is useful and may be invoked when a switch between the channel and the audiovisual input occurs.

In some embodiments, the decoder (100) traverses all frequency bands to see if there is a continuous period of relative silence (eg 250 ms configurable by the system and / or user). It can be configured to determine whether a reset event occurs by determining whether or not a large momentary drop in the excited level occurs.

In some embodiments, the decoder (100) responds to receiving an indicator (such as a reset event) provided in the encoded audio signal (102) by an upstream encoder (eg 150). It is configured to determine that a reset event has occurred.

The reset mechanism may be configured to issue a reset when the decoder (100) determines that a reset event has occurred. In some embodiments, the reset mechanism is configured to use a slightly more aggressive cut behavior of the DRC compression curve to prevent hard start (eg, loud programs / channels / audiovisual sources). Additional, optional or alternative, the decoder (100) may be configured to implement safeguards to gracefully recover when the decoder (100) detects that a reset was accidentally triggered. ..

<u style="single">&lt;17. Gain provided by the encoder&gt;</u> In some embodiments, the audio decoder gains (eg, audio data blocks, audio data frames, etc.) about individual parts of the audio content to be encoded in the encoded audio signal. It can be configured to compute one or more sets of (such as DRC gains). Their set of gains generated by the audio encoder is a single broadband (or wideband) for all channels (eg left front, right front, low frequency effect or LFE, center, left surround, right surround, etc.). First set of gains containing gains; Second set of gains containing individual broadband (or bandwidth) gains for individual subsets of channels; For individual subsets of channels and the first number (eg two) A third set of gains, including individual broadband (or broadband) gains for each of the individual bands (for example, two bands in each channel); for individual subsets of the channel and a second number. It may include a fourth set of gains, including individual broadband (or broadband) gains for each of the individual bands (eg, four bands in each channel). The subset of channels described herein may be one such as a subset containing left anterior, right anterior and LFE channels, a subset including a central channel, a subset containing left surround and right surround channels, and the like. Good.

In some embodiments, the audio encoder is about one or more parts of the audio content (eg, audio data blocks, audio data frames, etc.) and said one or more parts of the audio content. It is configured to transmit one or more sets of calculated gains in a time-synchronous manner. An audio decoder that receives said one or more pieces of audio content may select and apply a gained set of the gained sets of gains with little or no delay. it can. In some embodiments, the audio encoder carries said one or more sets of gains in one or more subframes (eg, using differential coding, etc.) as shown in FIG. Subframing techniques can be implemented. In one example, subframes may be encoded within an audio data block or audio data frame from which their gains are calculated. In another example, the subframes may be encoded within the audio data blocks or audio data frames that precede the audio data blocks or audio data frames whose gains are calculated. Good. In another non-limiting example, the subframes are within an audio data block or audio data frame within a period of time from the audio data block or audio data frame from which their gains are calculated. It may be encoded. In some embodiments, Huffman and differential coding may be used to populate and / or compress the subframes carrying those sets of gains.

<u style="single">&lt;18. Illustrative system and process flow&gt;</u> FIG. 5 shows an exemplary codec system in an exemplary embodiment without limitation. A content creator, which may be a processing unit within an audio encoder, such as the 150, is configured to provide audio content (audio) to an encoder unit (NGC encoder). The encoder unit formats the audio content into audio data blocks and / or frames and encodes the audio data blocks and / or frames into an encoded audio signal. Content Creator may include one or more dialog loudness levels (dialnorm) and one or more dynamic range compression curve identifiers (compression curve ID) for one or more programs, commercials, etc. in audio content. ") Is also configured to establish / generate. The content creator may determine the dialog loudness level from one or more dialog audio tracks in the audio content. The dynamic range compression curve identifier may be selected, at least in part, based on user input, system configuration parameters, and the like. The content creator may be a person (eg, an artist, an audio engineer, etc.) who uses the tools to generate the audio content and dialnorm.

Based on the dynamic range compression curve identifier, the encoder (150) sets the corresponding reference dialog loudness level (reference level) for multiple playback environments supported by the one or more dynamic range compression curves. Generate one or more DRC parameter sets that include, but are not limited to. These DRC parameter sets may be encoded in the metadata of the encoded audio signal, such as in-band with the audio content, out-of-band with the audio content, and so on. Operations such as compression, format multiplexing (MUX), etc. may be performed as part of producing an encoded audio signal that can be delivered to an audio decoder such as 100. Encoded audio signals have syntax that supports carrying audio data elements, DRC parameter sets, reference loudness levels, dynamic range compression curves, functions, look-up tables, Huffman codes used in compression, subframes, etc. It may be encoded. In some embodiments, the syntax allows an upstream device (eg, encoder, decoder, transcoder, etc.) to transmit gain to a downstream device (eg, decoder, transcoder, etc.). In some embodiments, the syntax used to encode the data into the encoded audio signal and / or decode the data from the encoded audio signal relies on the gain calculated by the upstream device. It is configured to support backward compatibility so that the device may optionally continue to do so.

In some embodiments, the encoder (150) computes two or more sets of gains for audio content (eg, gain smoothing with an appropriate reference dialog loudness level, DRC gains, etc.). These sets of gains may provide the one or more dynamic range compression curves in the metadata encoded in the audio signal encoded with the audio content. The first set of gains may correspond to broadband (or broadband) gains for all channels in the speaker configuration or profile (eg, default). A second set of gains may correspond to broadband (or broadband) gains for each of all channels in the speaker configuration or profile. A third set of gains may correspond to broadband (or broadband) gains for each of the two bands in each of all channels in the speaker configuration or profile. A fourth set of gains may correspond to broadband (or broadband) gains for each of the four bands in each of all channels in the speaker configuration or profile. In some embodiments, the set of gains calculated for a speaker configuration may be transmitted in the metadata along with a dynamic range compression curve (eg, parameterized) for that speaker configuration. In some embodiments, the set of gains calculated for a speaker configuration may replace the dynamic range compression curve (eg, parameterized) for that speaker configuration in the metadata. Additional speaker configurations or profiles may be supported under the techniques described in this article.

The decoder (100) is configured to extract audio data blocks and / or frames and metadata from the encoded audio signal through actions such as decompression, deformatting, and demultiplexing (DEMUX). Will be done. The extracted audio data blocks and / or frames may be decoded into audio data elements or samples by a decoder unit (NGC decoder). The decoder (100) also determines the profile in the decoder (100) for the particular playback environment in which the audio content is rendered, and selects a dynamic range compression curve from the metadata extracted from the encoded audio signal. It is configured to do. A digital audio processing unit (DAP) is configured to apply DRC or other behavior to an audio data element or sample for the purpose of producing an audio signal that drives an audio channel in a particular playback environment. To. The decoder (100) can calculate and apply the DRC gain based on the audio data block or frame and the selected dynamic range compression curve. The decoder (100) also outputs a dialog based on the reference dialog loudness level associated with the selected dynamic range compression curve and the dialog loudness level in the metadata extracted from the encoded audio signal. -You can also adjust the loudness level. The decoder (100) can then apply gain limiters specific to the audio content and playback scenarios associated with the particular playback environment. In this way, the decoder (100) can render / play the audio content as tailored to the playback scenario.

FIG. 5A shows another exemplary decoder (which may be the same as the decoder 100 in FIG. 5). As shown in Figure 5A, the decoder in Figure 5A decompresses, unformats, multiplexes (DEMUX) audio data blocks and / or frames and metadata from the encoded audio signal, for example. It is configured to extract through the operation of. The extracted audio data blocks and / or frames may be decoded into audio data elements or samples by the decoder unit (decode). The decoder of FIG. 5A is further configured to perform DRC gain calculations on a set of default gains based on a default compression curve, smoothing constants associated with the default compression curve, and the like. The decoder in Figure 5A also collects a set of differential gains for the non-default gain profile from the profile-related metadata in the metadata to produce a DRC-enhanced audio output that drives the audio channel in a particular playback environment. The set of non-differential gains for the non-default gain profile in the decoder of Figure 5A, which is extracted and the audio content is rendered, is determined and the set of non-differential gains and other behaviors are applied to the audio data element or sample. Configured to apply. The decoder in Figure 5A may or may not implement support for the decoder in Figure 5A itself to perform a set of gain generation operations to obtain a set of non-differential gains directly for the non-default gain profile. Audio content can be rendered / played according to the default gain profile.

Figures 6A through 6D show exemplary process flows. In some embodiments, one or more computing devices or units in a media processing system may perform this process flow.

Figure 6A shows an exemplary process flow that may be implemented by the audio decoders described in this article. In block 602 of FIG. 6A, a first device (such as the audio decoder 100 of FIG. 1A) receives an audio signal containing audio content and definition data for one or more dynamic range compression curves. To do.

At block 604, the first device determines a particular playback environment.

In block 606, the first device establishes a particular dynamic range compression curve for that particular playback environment based on the definition data for the one or more dynamic range compression curves extracted from the audio signal. To do.

At block 608, the first device performs one or more Dynamic Range Control (DRC) operations on one or more parts of the audio content extracted from the audio signal. The one or more DRC operations are based, at least in part, on one or more DRC gains obtained from a particular dynamic range compression curve.

In certain embodiments, the definition data for the one or more dynamic range compression curves is related to attack time, release time, or at least one of the one or more dynamic range compression curves. Including level.

In certain embodiments, the first apparatus is: the step of calculating one or more loudness levels for said one or more parts of audio content; the particular dynamic range compression curve and said audio content. It is further configured to perform steps such as determining the one or more DRC gains based on the one or more loudness levels for the one or more portions.

In certain embodiments, at least one of the loudness levels calculated for the one or more parts of the audio content is a specific loudness level, broadband range associated with one or more frequency bands. One or more of a cross-broadband loudness level, a wideband loudness level across a wideband range, a broadband loudness level across multiple frequency bands, a wideband loudness level across multiple frequency bands, and so on. ..

In certain embodiments, at least one of the loudness levels calculated for said one or more parts of the audio content is an instantaneous loudness level or a loudness level smoothed over one or more time intervals. One or more of.

In certain embodiments, the one or more actions relate to one or more such as adjusting the dialog loudness level, gain smoothing, gain limiting, dynamic equalization, noise compensation, and the like.

In certain embodiments, the first apparatus further: extracts one or more dialog loudness levels from the encoded audio signal; said one or more dialog loudness levels. It is configured to perform steps such as adjusting to the reference dialog loudness level;

In certain embodiments, the first device further: the step of extracting one or more auditory scene analysis (ASA) parameters from the encoded audio signal; in smoothing the gain applied to the audio content. A step of changing one or more time constants used, wherein the gain is related to one or more of the one or more DRC gains, such as performing gain smoothing or gain limiting. Is configured.

In certain embodiments, the first apparatus is further: a step of determining that a reset event occurs in said one or more parts of the audio content based on an index of the reset event of the reset. The indicators are extracted from the encoded audio signal, and in response to determining that the reset event occurs in said one or more parts of the audio content. A step of performing one or more actions on one or more gain smoothing actions being performed at the time of determining that the reset event occurs in the one or more parts of the content; etc. Configured to run.

In one embodiment, the first apparatus is further: a stage of maintaining a histogram of instantaneous loudness levels, which contains the instantaneous loudness levels calculated from a time interval in the audio content. , Stage; A step of determining whether a specific loudness level is above a threshold in the high probability region of the histogram, the specific loudness level being calculated from a portion of the audio content. , Step; Determine that a loudness transition is occurring and speed up the loudness transition in response to determining that the specific loudness level is above the threshold in the high probability region of the histogram. It is configured to perform steps such as shortening the time constant used in gain smoothing to do so.

Figure 6B shows an exemplary process flow that may be implemented by the audio encoders described in this article. At block 652 of FIG. 6B, a second device (such as the audio encoder 150 of FIG. 1B) receives the audio content in the source audio format.

At block 654, the second device acquires definition data for one or more dynamic range compression curves.

At block 656, the second device produces an audio signal that includes said audio content and said defined data about the one or more dynamic range compression curves.

In certain embodiments, the second apparatus further comprises determining one or more identifiers for the one or more dynamic range compression curves; a reference data reservoir based on the one or more identifiers. It is configured to perform steps such as retrieving the definition data for the one or more dynamic range compression curves from.

In certain embodiments, the second device further: with the step of calculating one or more dialog loudness levels for said one or more parts of the audio content; said one or more dialogs. The loudness level is configured to perform, for example, the step of encoding into the encoded audio signal together with the one or more parts of the audio content.

In certain embodiments, the second device is: performing an auditory event scene (ASA) on said one or more parts of the audio content; said one or more parts of the audio content. The stage of generating one or more ASA parameters based on the results of the ASA for; the one or more ASA parameters, along with the one or more parts of the audio content, said encoded audio. It is configured to perform the steps of encoding into the signal; etc.

In certain embodiments, the second device further: a step of determining that one or more reset events occur in said one or more parts of the audio content; said one or more resets. A step of encoding one or more indicators of an event into the encoded audio signal together with the one or more parts of the audio content; etc. is configured to perform.

In certain embodiments, the second device is further configured to encode said one or more parts of the audio content into one or more audio data frames or blocks of audio data.

In certain embodiments, the first DRC gain of the one or more DRC gains is for each channel in the first true subset of all channels in a particular speaker configuration for that particular playback environment. This is true, on the other hand, the second different DRC gain of the one or more DRC gains is each in the second true subset of the set of all channels in the particular speaker configuration corresponding to that particular playback environment. Applies to the channel.

In certain embodiments, the first DRC gain of the one or more DRC gains applies to the first frequency band and the second different DRC gain of the one or more DRC gains is the second different frequency band. Applies to.

In certain embodiments, the one or more portions of the audio content include one or more of an audio data frame or audio data block. In certain embodiments, the encoded audio signal is part of an audiovisual signal.

In certain embodiments, the one or more DRC gains are defined in the loudness region.

Figure 6C shows an exemplary process flow that may be implemented by the audio decoders described in this article. In block 662 of FIG. 6C, a third device (eg, audio decoder 100 of FIG. 1A, audio decoder of FIG. 5, audio decoder of FIG. 5A, etc.) has audio content and one or more dynamics. Receives an audio signal containing definition data for a range compression (DRC) curve and one or more sets of differential gains.

In block 664, the third device identifies a particular set of differential gains for a gain profile in a particular reproduction environment, out of one or more sets of differential gains. The third device also identifies, among the one or more DRC curves, the default DRC curve associated with the particular set of differential gains.

At block 666, the third device produces a set of default gains, at least in part, based on the default DRC curve.

In block 668, the third device is one or more of the audio content extracted from the audio signal, at least in part, based on the combination of the set of default gains and the particular set of differential gains. Perform one or more actions on the part of.

In certain embodiments, the set of default gains includes, at least in part, the non-differential gains generated by performing a set of gain generation operations based on the default DRC curve.

In certain embodiments, the default DRC curve represents a default gain profile. In certain embodiments, the particular set of differential gains in relation to the default DRC curve represents a non-default gain profile. In certain embodiments, the audio signal does not include definition data for the non-default DRC curve corresponding to the non-default gain profile.

In certain embodiments, the particular set of differential gains is the set of non-default non-default gains generated for the non-default gain profile and the non-difference generated for the default gain profile represented by the default DRC curve. Includes the gain difference from the set of default gains. The set of non-difference non-default gains and the set of non-difference default gains may be generated by an upstream audio decoder that encodes the audio signal.

In certain embodiments, at least one set of the non-differential non-default gains or a set of the non-differential default gains is not provided as part of the audio signal.

Figure 6D shows an exemplary process flow that may be implemented by the audio decoders described in this article. In block 672 of FIG. 6D, a fourth device (eg, audio encoder 150 of FIG. 1A, audio encoder of FIG. 5, etc.) receives audio content in the source audio format .

At block 674, the fourth device, at least in part, produces a set of default gains based on a default dynamic range compression (DRC) curve that represents the default gain profile.

In block 676, the fourth device produces a set of non-default gains for the non-default gain profile.

In block 678, the fourth device produces a set of differential gains, at least in part, based on the set of default gains and the set of non-default gains. The set of differential gains represents the non-default gain profile in relation to the default DRC curve.

At block 680, the fourth device produces an audio signal containing said audio content and said defined data for one or more DRC curves and for one or more sets of differential gains. The one or more sets of differential gains include said sets of differential gains.

In some embodiments, the non-default gain profile is represented by a DRC curve. In certain embodiments, the audio signal does not include definition data for the DRC curve representing the non-default gain profile. In some embodiments, the non-default gain profile is not represented by a DRC curve.

In certain embodiments, a device having a processor and configured to perform any of the methods described herein.

In certain embodiments, a non-transitory computer-readable storage medium containing software instructions that, when executed by one or more processors, triggers the execution of any of the methods described herein. It should be noted that although separate embodiments are discussed in this article, any combination of embodiments and / or partial embodiments discussed in this article may be combined to form further embodiments. ..

<u style="single">&lt;19. Mounting mechanism-Overview of hardware&gt;</u> According to certain embodiments, the techniques described herein are implemented by one or more special purpose computing devices. The special purpose computing device may be fixedly configured to perform the technique, or may be one or more application specific integrated circuits (ASICs) or field programmable gate arrays (FPGAs). It may include a digital electronic device that is persistently programmed to perform the technique, or one that is programmed to perform the technique according to program instructions in firmware, memory, or other storage or combination. Alternatively, it may include multiple general purpose hardware processors. Such special purpose computing devices may combine custom fixed configuration logic, ASICs or FPGAs with custom programming to accomplish this technique. A special purpose computing device is a desktop computer system, a portable computer system, a handheld device, a networking device, or any other device that incorporates fixed configurations and / or program logic to implement the technique. May be good.

For example, FIG. 7 is a block diagram showing a computer system 700 in which certain embodiments of the present invention can be implemented. The computer system 700 includes a bus 702 or other communication mechanism for communicating information and a hardware processor 704 coupled to the bus 702 for processing information. The hardware processor 704 may be, for example, a general purpose microprocessor.

Computer system 700 is main memory coupled to bus 702 to store information and instructions to be executed by processor 704, such as random access memory (RAM) or other dynamic storage. Also includes 706. Main memory 706 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by processor 704. When such an instruction is stored on a non-temporary storage medium accessible to the processor 704, it makes the computer system 700 a device-specific special purpose machine to perform the processing specified in the instruction. ..

Computer system 700 further includes read-only memory (ROM) 708 or other static storage device for storing static information and instructions for processor 704 coupled to bus 702. A storage device 710, such as a magnetic disk or optical disk, is provided and coupled to bus 702 to store information and instructions.

The computer system 700 may be coupled via a bus 702 to a display 712, such as a liquid crystal display (LCD), for displaying information to the computer user. An input device 714 containing alphanumeric characters and other keys is coupled to bus 702 to convey information and command selection to processor 704. Another type of user input device is a cursor control 716, such as a mouse, trackball, or cursor direction key, to convey direction information and command selection to the processor 704 and to control cursor movement on the display 712. is there. The input device typically has two degrees of freedom in two axial directions, the first axis (eg x) and the second axis (eg y), which allows the device to specify its position in the plane. ..

The computer system 700 combines device-specific fixed configuration logic, one or more ASICs or firmwares, and computer systems to turn the computer system 700 into a special purpose machine to perform the techniques described in this article. You may use firmware and / or program logic to and / or program. According to one embodiment, the techniques of this paper are performed by computer system 700 in response to processor 704 executing one or more sequences of one or more instructions contained in main memory 706. Will be done. Such instructions may be read into main memory 706 from another storage medium such as storage device 710. By executing a sequence of instructions contained in main memory 706, processor 704 executes the process steps described in this article. In alternative embodiments, fixed configuration circuits may be used in place of or in combination with software instructions.

As used herein, the term "storage medium" refers to any non-temporary medium that stores data and / or instructions that cause a machine to operate in a particular way. Such storage media may include non-volatile media and / or volatile media. Non-volatile media include, for example, optical or magnetic disks such as storage device 710. Volatile media include dynamic memory such as main memory 706. Common forms of storage media include, for example, floppy disks, flexible disks, hard disks, semiconductor drives, magnetic tape or any other magnetic data storage medium, CD-ROM, any other optical data storage medium, holes. Includes any physical medium with a pattern, RAM, PROM and EPROM, flash EPROM, NVRAM, and any other memory chip or cartridge.

The storage medium is different from the transmission medium, but may be used in connection with the transmission medium. The transmission medium participates in transferring information between storage media. For example, the transmission medium includes coaxial cable, copper wire and optical fiber, and includes wires forming bus 702. The transmission medium can also take the form of acoustic or light waves as produced during radio and infrared data communications.

Various forms of media can be involved in delivering one or more sequences of one or more instructions to processor 704 for execution. For example, the instructions may initially be carried on the magnetic disk or semiconductor drive of the remote computer. The remote computer can load the instruction into its dynamic memory and send the instruction using a modem over the telephone line. A modem local to computer system 700 can receive data on the telephone line and use an infrared transmitter to convert that data into an infrared signal. The infrared detector can receive the data carried in the infrared signal and a suitable circuit can put that data on the bus 702. The bus 702 transports the data to the main memory 706, and the processor 704 fetches and executes an instruction from the main memory 706. Instructions received by main memory 706 may optionally be stored on storage device 710 before or after execution by processor 704.

The computer system 700 also includes a communication interface 718 coupled to bus 702. Communication interface 718 provides a bidirectional data communication coupling to network link 720 connected to local network 722. For example, the communication interface 718 may be an integrated services digital network (ISDN) card, cable modem, satellite modem or modem for providing a data communication connection to a corresponding type of telephone line. As another example, the communication interface 718 may be a local area network (LAN) card for providing a data communication connection to a compatible LAN. Wireless links may also be implemented. In any such implementation, the communication interface 718 sends and receives electrical, electromagnetic or optical signals that carry digital data streams that represent different types of information.

Network link 720 typically provides data communication to other data devices over one or more networks. For example, network link 720 may provide a connection through local network 722 to host computer 724 or data equipment operated by Internet Service Provider (ISP) 726. ISP 726 provides data communication services through the global packet data communication network, now commonly referred to as the "Internet" 728. Both local networks 722 and Internet 728 use electrical, electromagnetic or optical signals that carry digital data streams. Signals through various networks and signals over network link 720 and through communication interface 718 that carry digital data to and from computer system 700 are exemplary forms of transmission medium.

The computer system 700 can send messages and receive data, including program code, through the network (s), network link 720, and communication interface 718. In the Internet example, server 730 may send the requested code for an application program through Internet 728, ISP 726, local network 722 and communication interface 718.

The received code may be executed by processor 704 as it is received and / or stored in storage device 710 or other non-volatile memory for later execution.

<u style="single">&lt;20. Equivalents, extensions, alternatives, etc.&gt;</u> In the above specification, exemplary embodiments of the present invention have been described with reference to a number of individual details that may vary from implementation to implementation. Thus, the only and exclusive indicator of what is the invention and what is intended to be the invention by the applicant is in the claims of the patent granted to this application. Such claims, including any subsequent corrections, are of a particular patented form. If there is a definition explicitly stated in this article for the terms contained in such claims, it governs the meaning of the terms used in the claims. Therefore, any limitation, element, attribute, feature, advantage or characteristic not explicitly stated in the claims should not limit the scope of such claim in any way. Therefore, specifications and drawings should be viewed in an exemplary sense rather than in a restrictive sense.

Some aspects are described. [Aspect 1] A step of receiving an audio signal including audio content and one or more sets of differential gains; a gain profile in the particular playback environment of the one or more sets of differential gains. And the step of identifying a particular set of differential gains for; and the step of generating a set of default gains based at least on the default dynamic range compression (DRC) curve associated with that particular set of differential gains; One or more for one or more parts of the audio content extracted from the audio signal, at least in part, based on a combination of said set of default gains and said particular set of differential gains. A method performed by one or more computers, including steps to perform the operation. [Aspect 2] The method of aspect 1, wherein the set of default gains comprises, at least in part, a non-differential gain generated by performing a set of gain generation operations based on the default DRC curve. [Aspect 3] The method according to aspect 1 or 2, wherein the default DRC curve represents a default gain profile. [Aspect 4] The method according to any one of aspects 1 to 3, wherein the particular set of differential gains in relation to the default DRC curve represents a non-default gain profile. [Aspect 5] The method according to aspect 4, wherein the audio signal does not include definition data for a non-default DRC curve corresponding to the non-default gain profile. [Aspect 6] The particular set of differential gains includes a set of non-default non-default gains generated for the non-default gain profile and a set of non-differential default gains generated for the default gain profile represented by the default DRC curve. The method according to any one of aspects 1 to 5, comprising a gain difference between. [Aspect 7] The method according to aspect 6, wherein the set of non-difference non-default gains and the set of non-difference default gains are generated by an upstream audio decoder that encodes the audio signal. [Aspect 8] The method of aspect 6, wherein at least one of the set of non-differential non-default gains or the set of non-differential default gains is not provided as part of the audio signal. [Aspect 9] One of the attack time, release time or reference loudness level in which the definition data for the one or more DRC curves is related to at least one of the one or more DRC curves. The method according to any one of aspects 1 to 8, which comprises one or more. [Aspect 10] 9. The method of aspect 9, wherein the reference loudness level represents a targeted range of playback levels for rendering the audio content by an audio decoder. [Aspect 11] A step of calculating one or more loudness levels for the one or more parts of the audio content; based on the set of non-differential default gains and the particular set of differential gains. With the stage of generating a set of non-differential non-default gains; The method according to any one of aspects 1 to 10, further comprising applying the set of non-differential non-default gains to said one or more parts of the audio content. [Aspect 12] At least one of the one or more loudness levels calculated for the one or more parts of the audio content is a specific loudness level, broadband range for one or more frequency bands. A method according to aspect 11, wherein the broadband loudness level spans, the broadband loudness level over a wide band range, the broadband loudness level over a plurality of frequency ranges, or one or more of the broadband loudness levels over a plurality of frequency ranges. .. [Aspect 13] At least one of the one or more loudness levels calculated for the one or more parts of the audio content is smoothed over an instantaneous loudness level or one or more time intervals. The method of aspect 11, wherein the loudness level is one or more. [Aspect 14] The one or more actions include adjusting the dialog loudness level, one or more actions related to one or more of gain smoothing, gain limiting, dynamic equalization or noise compensation. , The method according to any one of aspects 1 to 13. [Aspect 15] The method according to any one of aspects 1 to 14, wherein the method is performed by an audio decoding device and the default DRC curve is defined in the audio decoding device. [Aspect 16] The stage of receiving definition data for one or more Dynamic Range Compression (DRC) curves; the default DRC of the one or more DRC curves associated with the particular set of differential gains. The method according to any one of aspects 1 to 15, further comprising the step of identifying a curve. [Aspect 17] A step of extracting one or more auditory scene analysis (ASA) parameters from the encoded audio signal; The method of any one of aspects 1-16, further comprising a step of varying one or more time constants used in smoothing the gain applied to the audio content. [Aspect 18] A step of determining that a reset event occurs in the one or more parts of the audio content based on an index of the reset event, wherein the index of the reset is encoded. Extracted from the audio signal, with stages; In response to determining that the reset event occurs in the one or more parts of the audio content, the reset event occurs in the one or more parts of the audio content. The method according to any one of aspects 1 to 17, further comprising performing one or more actions on one or more gain smoothing operations being performed at the time of determination. [Aspect 19] At least one of the one or more smoothing operations uses the first smoothing time constant before the reset event, and at least one of the one or more smoothing operations The method of aspect 18, wherein a second smoothing time constant smaller than the first smoothing time constant is used in response to determining that the reset event occurs. [Aspect 20] A step of maintaining a histogram of the instantaneous loudness level, wherein the histogram contains the instantaneous loudness level calculated from a certain time interval in the audio content. A step of determining whether a specific loudness level is below a threshold in the high probability region of the histogram, wherein the specific loudness level is calculated from a portion of the audio content. In response to determining that the specific loudness level is below the threshold in the high probability region of the histogram: The item according to any one of aspects 1 to 19, further comprising a step of determining that a loudness transition has occurred and shortening the time constant used in gain smoothing to speed up the loudness transition. Method. [Aspect 21] The specific set of differential gains includes the first differential gain associated with each channel in the first true subset of the set of all channels in a particular speaker configuration, and the specific set of differential gains. The method of any one of aspects 1-20, wherein the set comprises a second differential gain associated with each channel in the second true subset of the set of all channels in the particular speaker configuration. .. [Aspect 22] The specific set of differential gains includes a first differential gain related to a first frequency band, and the particular set of differential gains includes a second different difference related to a second different frequency band. The method according to any one of aspects 1 to 21, comprising a gain. [Aspect 23] The method according to any one of aspects 1 to 22, wherein the one or more parts of the audio content comprises one or more of an audio data frame, an audio data block or an audio sample. [Aspect 24] The method according to any one of aspects 1 to 23, wherein the particular set of differential gains is defined in a loudness region. [Aspect 25] The method according to any one of aspects 1 to 24, wherein the encoded audio signal is a part of an audiovisual signal. [Aspect 26] Receiving audio content in the source audio format; at least in part, generating a set of default gains based on a default dynamic range compression (DRC) curve, said default DRC. The curve represents the default gain profile, with the steps; and with the steps that generate a set of non-default gains for the non-default gain profile; At least in part, at the stage of generating the set of differential gains based on the set of default gains and the set of non-default gains, the set of differential gains in relation to the default DRC curve. A step representing the non-default gain profile; one or one including a step of generating an audio signal containing said audio content and said one or more sets of differential gains including said set of differential gains. A method performed by multiple computing devices. [Aspect 27] The method of aspect 26, wherein the non-default gain profile is represented by a DRC curve. 28. The method of aspect 27, wherein the audio signal does not include definitional data for the DRC curve representing the non-default gain profile. [Aspect 29] The method according to any one of aspects 26 to 28, wherein the non-default gain profile is not represented by a DRC curve. [Aspect 30] A step of determining one or more identifiers for the one or more dynamic range compression curves; 28. One of embodiments 26-29, further comprising retrieving the defined data for the one or more dynamic range compression curves from the reference data reservoir based on the one or more identifiers. Method. [Aspect 31] The set of default gains comprises, at least in part, a first non-differential gain generated by performing a first set of gain-generating operations based on the default DRC curve. 12. One of aspects 26-30, wherein the set of default gains comprises a second non-differential gain generated by performing a second set of gain generation operations for the non-default gain profile. Method. [Aspect 32] A step of calculating one or more dialog loudness levels for one or more parts of the audio content; Aspects 26-31, further comprising the step of encoding the one or more dialog loudness levels into the encoded audio signal together with the one or more parts of the audio content. The method described in any one of them. [Aspect 33] The method of aspect 32, wherein at least one of the one or more dialog loudness levels is determined from one or more audio tracks that include dialog audio content. [Aspect 34] A step of performing an auditory scene analysis (ASA) on the one or more parts of the audio content; based on the results of the ASA on the one or more parts of the audio content. With the stage of generating one or more ASA parameters; Any of aspects 26-33, further comprising the step of encoding the one or more ASA parameters together with the one or more parts of the audio content into the encoded audio signal. The method described in paragraph 1. [Aspect 35] A step of determining that one or more reset events occur in one or more parts of the audio content; Further including the step of encoding one or more indicators of the one or more reset events together with the one or more parts of the audio content into the encoded audio signal. The method according to any one of aspects 26 to 34. [Aspect 36] Any one of aspects 26 to 35, further comprising the step of encoding one or more parts of the audio content into one or more audio data frames or blocks of audio data. Item description method. [Aspect 37] The method according to any one of aspects 26 to 36, wherein at least one of the one or more dynamic range compression curves is defined in a loudness region. [Aspect 38] The method according to any one of aspects 26 to 37, wherein the encoded audio signal is a part of an audiovisual signal. [Aspect 39] The definition data for the one or more dynamic range compression curves includes one or more sets of parameters, and at least one set in the one or more sets of parameters is a look-up table, curve or plural. The method according to any one of aspects 26 to 38, which represents one or more of the segment division straight lines. 40. The method of any one of aspects 26-39, wherein the encoded audio signal comprises an index for selecting the DRC curve defined in the receiving device as the default DRC curve. [Aspect 41] A step of sending definition data for various DRC curves in the encoded audio signal; The method according to any one of aspects 26 to 40, further comprising a step of including an index for selecting the default DRC curve among the one or more DRC curves. [Aspect 42] A media processing system configured to perform the method according to any one of aspects 1 to 41. [Aspect 43] A device having a processor configured to perform the method according to any one of aspects 1 to 41. [Aspect 44] A non-temporary computer-readable storage medium containing software instructions that, when executed by one or more processors, triggers the execution of the method according to any one of aspects 1-41.