Method and device for the reproduction of a binaural output signal which is derived from a monaural input signal

0001The invention relates to a method for reproducing one of a monaural input signal generated from a first Output signal and a second output current binaural output signal and an apparatus for performing the method according to the preamble of claim 1 or of claim. 8

0002In the modern world of communications increasingly be intelligent Communication devices, such. As PCs or PDAs, for Voice communication used, with a combination of these Communication devices for example via VoIP.

0003The packet-based communication using VoIP and the causes related use of so-called VoIP codecs adverse influences on the voice quality. For example with medium to higher maturities in signal transmission expected to give audible echoes can arise. additionally just need to participate in a packet-based communication Discards be reckoned with maturities often higher and its damping is lower than in a natural Environment is to be found. Therefore, measures to suppress disturbing echoes provide, preferably by the use of echo cancellers in the data terminals.

0004Echo Canceller based on common standards, such as ITU-T G.168 (2002), where, inter alia, gateway transitions into the classic Telephone network are discussed. For VoIP terminals alternatively ITU-T G.165 (1993) are used, wherein this respect. Echo and dispersion to be achieved suppression much milder specify requirements than do standards traditional telephony is the case.

0005If the data terminal itself is designed as a VoIP terminal, have they over dedicated VoIP terminals the disadvantages higher maturities in the signal transmission and the lack of Echo Canceller on. In particular, due to the lack canceller need for a packet-based communication of this Type Headsets (headsets) can be used.

0006However, conventional binaural headphones produce a more unnatural auditory event, because of the sound from the head and from the Outer ear is no longer affected. In natural hearing received both ears, the signals of all sound sources, so that delays, level differences and differences in sound create a spatial listening experience. investigations to the direction of perception eintreffendem sound show that interaural delay and level differences only in terms a horizontal plane of symmetry of the head relevant are, so that here the direction of the incoming can be determined sound. With respect to a vertical plane of symmetry the head occur any time delays or However, level differences on, here is a clear sense of directions the incoming sound on sound differences determine. Three-dimensional listening is decisive for the spatial orientation, the separation of different Sound sources (see Blauert, Jens (June 1997). Spatial Hearing, MIT Press, Chap. 5.3) and the suppression of perception discards (ibid., Sec. 5.4), since the sound sources when using headphones directly on the ears are, any three-dimensional listening is inhibited. The right ear receives namely exclusively the signals the right speaker, whereas the left ear only the signals of the left speaker receives.

0007The invention is thus based on the object, a method and a device for reproducing a monaural from a Input signal output signal generated to the effect further develop the quality of means Headsets unrealized monaural VoIP voice connections is improved.

0008This object is achieved by a method according to claim 1 or achieved by a device according to claim eighth

0009According to the invention the object is achieved by a method, wherein a signal generated from a monaural input signal, of a first output signal and a second output signal existing binaural output signal over at least a first and a second speaker binaural Headset especially for VoIP applications is reproduced. Here, from the monaural input signal by Phasenverschiebung- and / or amplitude gain the first output signal and / or the second output signal produced for the binaural simulation to an auditory to obtain that a perceived static and / or represents dynamic positioning of a sound event.

0010Furthermore, the object is solved by an apparatus in which in particular for a binaural headset VoIP applications at least a first and a second Speaker to output a monaural from an input signal produced, first from an output signal and a second output existing binaural output signal and a connection to a receiver-side data terminal having. A signal processing means generates from the monaural input signal by phase shift and / or Amplitude gain the first output signal and / or the second output signal for binaural simulation to an auditory to obtain that a perceived static and / or dynamic positioning of a sound event represents.

0011An essential point of the invention lies in the fact that, due to binaural simulation a spacious, largely as naturally perceived hearing despite using headphones is achieved.

0012Here, with the help of phase, delay, Dämpfungsund / or HRTF (Head Related Transfer Function) -Verarbeitungselementen the natural way of the sound simulated, namely the free-field, the outer ear and the ear canal transmission, or the basis of phase differences, time delays, Level differences and sound differences caused natural Hear. Due to this reproduction, it is possible perception discards, such as coloration or Echoes largely suppress, especially the occurrence of Discards is controlled mentally to a certain degree and among other things, depends on experience and attention. This stirred particular the fact that concurrent Sound events, the different sound sources originate, are easily distinguishable. Thereby, the ability improves the listener, to a sound source to concentrate and their sound events over the Sound events of other sources emphasize perceptually. In addition, by simulating the three-dimensional Hearing the precedence effect, ie the law of the first wavefront available as soon as the sound of several coherent Sources from different directions on the listener applies. The sound event seems, then only a single to get direction, echoes are not perceived.

0013It is thus provided in a first preferred embodiment, that the monaural input signal from a transmitting around / or a receiver-side data terminal of the VoIP application is supplied. This has the particular advantage, that produced next to the transmitter-side data terminal Sound event also from the receiver-side data terminal included sound event generated in the binaural simulation becomes. Even in natural hearing is your own voice audible as three-dimensional sound event, so that a clear demarcation to another sound source, eg. as a other speakers, is possible.

0014It is advantageously provided that the phase shift in a first part of the static positioning function the noise produced by the transmitter-side data terminal Sound event is simulated. For this purpose, the first output signal by a delay of the transmitter-side data terminal delivered input signal generated or made sign inversion and this the first loudspeaker supplied. Through the unchanged reproduction of the input signal is also the second output signal generated and this is supplied to the second speaker. Preferably is hereby static positioning of the transmitter side Terminal caused sound event "closer" to the reached second speaker. Due to the phase shift and the associated varying maturities the two output signals is here advantageously a first component for generating a three-dimensional auditory event implemented.

0015In an advantageous embodiment, the dynamic Positioning the noise produced by the transmitter-side data terminal Sound event in a second partial function simulates. These first carried a medium level comparison between the input signal delivered from the transmitter side data terminal and supplied by the receiver-side data terminal monaural input signal. Subsequently, the delivered by the transmitter-side data terminal input delayed to this first delay the first output signal to create. A second delay of the input signal provides the second output signal. The first Output signal is applied to the first speaker, the second Output signal is supplied to the second speaker. hereby is the dynamic positioning of the transmitter side Terminal caused sound event "closer" to achieved each speaker, the appropriate the Output due to a different duration first reached. With regard to the dynamic positioning of sound events is due to the phase shift and the associated varying maturities of two output signals advantageously a further component to generate a three-dimensional auditory event implemented.

0016With static and dynamic positioning here is the Simulation of directional perception of the incoming sound from the perspective of the receiver-side data terminal and the receiver side Participant described. In other words, it is simulated that the sound event produced from an specific direction arrives. If a static positioning simulates the processing of supplied takes place Sound such that the acoustic event produced by him a "Unmoved" transmitter-side participants suggest. The Simulation of a moving transmitter-end subscriber contrast describes its dynamic positioning. The sound is processed such that a change of location of the transmitter side is participant simulated. Both the simulation of the static and the dynamic positioning of the sound event therefore allow a in the case of audio transmission the course listening reinvented listening experience.

0017Preferably, in a third part-function static Positioning of the receiver-side data terminal caused sound event simulated. For this purpose there is a Delay supplied by the receiver-side data terminal monaural input signal to this first output signal reproduce. At the same time the input signal is Passed unchanged to it as a second output signal to provide. enters the first output signal then the second speaker, while the second output signal is supplied to the first speaker. The static positioning is thus realized, that caused by the receiver-side data terminal Sound event "closer" to the first speaker appears.

0018Own discards as proposed here with low delay, are desirable and in traditional telephony described in detail. See z. B. ITU-T G.131 (1996) and ITU-T G.111 (1993) Annex A, keyword STMR (Side Tone Masking Rating, Talkers's sidetone).

0019Advantageously, the static positioning of the transmitter-side data terminal caused sound event and the static positioning of the receiver side Terminal caused sound event simulated simultaneously. This essentially corresponds to a combination the first and the third sub-function. The incident Sound both at the voice terminals involved is thus perceptible from different directions, including the echo of the receiver-side terminal. At the same time a reinforcement of the will Präzedenzeffektes by the receiver-side data terminal sound produced reached. According to this, in Fig. 1 of the so-called echo Threshold drawn according Blauert. See Figure 3.13 of ITU-T G.131 for a typical gain in the terminal. clearly is the "profit" to TelR (Talker echo loudness rating) to recognize.

0020The inventive solution provides, in another embodiment before, the dynamic positioning of the transmitter side Terminal caused sound event and the static positioning of the receiver-side data terminal caused sound event to simulate the same. This essentially corresponds to a combination of second and the third part-function. Advantageously, thereby causing the receiver-side data terminal Sound event, the echo sound event and the transmitter-side data terminal caused sonic event different directions perceived. This makes it possible the incident, from the transmitter-side data terminal Sound and the incoming data from the receiver-side terminal Sound compared to the echo of the receiver side Terminal incoming sound to emphasize perceptually.

0021In a further preferred embodiment, the binaural Headset with a signal processing device formed that has at least a run-time member. The Runtime element accordingly produces the above-described Phase shift of the respective outputs. alternative or additionally, the signal processing means at least one damping element and / or at least one HRTF (Head Related Transfer Function) -Verarbeitungselement provide. In addition to the phase shifts then and amplitude gains and / or differences in sound produced. Advantageously, by means of these elements, with the Combination of elements and in particular with the combination all elements a realistic three-dimensional listening produced even when using binaural headphones because a natural hearing by time delays, intensity differences and colorations is characterized.

0022Further features and advantages of the inventive devices result from the features and benefits of The method according to the invention.

0023The invention is based on an embodiment described in more detail, the explained with reference to Figure becomes. In the figures:<sl><li>FIG. 1 is a graph showing the acceptance of the assessment conversation (Talker echo Tolerance Curves)</li><li>Fig. 2 shows an embodiment of the invention.</li></sl>

0024Fig. 1 shows so-called "Talker Echo Tolerance Curves", the close regard. vorzufindenden the echoes on voice quality leave. The curves permit accordingly acceptance assessment the conversation. Here, on the abscissa, medium echo transmission time (Mean Echo Transmission Time) T and the ordinate the Talker Echo Loudness Rating TelR applied. Curve K1 shows the hearing threshold (Masked Threshold), the curve K2 shows the Acceptable (Acceptable). The acceptable is equivalent to the curve in which an interfering Echo with a probability of 1% occurs. The Curve K3 represents the limiting case (Limiting Case), the curve K4 contrast binaural limit (Binaural Limiting Case) stereophonic speakers in an arrangement with an angle of 80 °).

0025In FIG. 2 is an embodiment of the invention Device shown as a functional block diagram. Here, a transmitter-side terminal with the reference numeral B, and a receiver side terminal by the reference numeral A named. The receiver-side data terminal A is ideally equipped with a binaural headset, in turn, a first speaker and a second L Speaker A has.

0026To control the signal flow accordingly, is between the respective terminals A, B, a signal processing device 1. In this embodiment, the signal processing means 1, three functional blocks F1, F2, F3 and a level processing element PVE on.

0027The functional blocks F1, F2 and F3 respectively have at least a delay element (not shown). Alternatively or additionally, can the functional blocks F1, F2 and F3 respectively and with at least one damping element and / or an HRTF (Head Related Transfer Function) -Verarbeitungselement (not shown) may be formed.

0028In this embodiment, function block F1 and Function block F2 connected in series, the function block F2 is connected in parallel to function block F1.

0029Of the transmitter-side terminal B is a voice connection set up to a receiving-side terminal A, where the linkage with the aid of a switching network over VoIP runs.

0030The transmitter-side Datendendgerät B transmits a monaural Input signal in a step 100 to the first function block F1. At the same time transmits the transmitter-side data terminal B the monaural input signal, in a step 101 to the function block F2 and in a step 102 to the level comparing element PVE.

0031Function block F1 delays the received signal and transmits it in a step 200 to the function block F3. At the same time can function block F1, the received signal pass unchanged and transmits the signal unchanged in a step 201, also to the function block F3. The subject to function block F2 applied from step 101 signal in the function block F2 of a first delay and comes with this, in a step 300 to function block F3. At the same subject from the step 101 to function block F2 signal applied to a second delay and is with this in a step 301 to function block F3 to hand over.

0032Also, the level comparison element PVE receives in a step 102 supplied by the transmitter-side data terminal B signal. At the same time is at the level comparison element PVE one from receiver-side data terminal A supplied to signal is passed in a step 502nd The above-described, performed in function block F2 first and second delay the delivered by the transmitter-side data terminal B Signal is then a function of a mean level comparison the signals supplied from the data terminals A, B.

0033At function block F3 are now available from the steps 200 and 300 or the steps 201 and 301 originating at signals. At the same time is to function block F3 consisting of a step 501 derived signal to the receiver-side data terminal. Imports coming from the steps 200 and 300 signals freely in this embodiment, function block F3 pass and then in step 400, the first loudspeaker L supplied. Also consisting of steps 201 and 301 resulting, present at function block F3 Signals the last function block F3 without further Machining pass, however, in a step 401 the second speaker R fed. Due to the already carried out in advance in the functional blocks F1 and F2 Signal delays takes place on the one hand a static positioning one induced by the transmission-side terminal B Sound event "closer" to the second speaker R while On the other hand, a dynamic positioning of the transmitter-side data terminal B induced sound event "Closer" is achieved at the respective loudspeaker, which in each case the signals with lower delays receives.

0034Function block F3 delays the pass in step 501 Signal and supplies it to the second speaker R. Simultaneously, the information transmitted in step 501 signal passes the function block F3 freely and is the first Speaker L passed. Thus, as mentioned above, Static positioning of the receiver side Terminal A induced sound event "closer" to the first loudspeaker L achieved.

0035Finally, the receiver-side data terminal A sends in a step 500 a signal without further processing directly at the receiver-side data terminal B.

0036The proposed resolution of a monaural input signal and its processing to obtain run-time differences allows a three-dimensional listening through a binaural headset that resembles a natural hearing is. Since natural hearing on transit time differences, Level differences and colorations of the incident resulting sound from different sound sources, , a three-dimensionally perceived hearing ideally be inspired by the fact that both run-time differences produced and level differences and colorations will.

0037The embodiment described above describes the Functional blocks as signal processing blocks, which aims is, time differences and thus phase differences from a to generate monaural input signal by its splitting. Alternatively, it is possible, the runtime elements by to replace damping elements. It is a spatial hearing modeled, the amplitude gain only about a or-damping is realized. It is also possible, exclusively HRTF (Head Related Transfer Function) -Verarbeitungselemente provide to the nature of the head and Ears and thus simulate the directivity of the ear. However, the functional blocks F1 to F3 can also all Signal processing elements simultaneously record, so a optimal results with respect. simulating the natural hearing to reach.

0038Alternatively, but not shown, it is possible, for example, the functional blocks F1 and F3 to combine. This matches with substantially in accordance with the embodiment illustrated Fig. 2, but without the from the sender side data terminal B supplied monaural input signal at the function block F2 to provide. The signals then pass through function block F3 with simultaneous processing of the receiver side A data terminal supplied input signal, to then be supplied to the loudspeakers L and R.

0039It is also possible, however, also not shown, the Function blocks F2 and F3 to be combined. Also herein may be substantially the already described FIG. 2 be based, however, in the absence of functional block F1. Delivered from the transmission-side terminal B monaural input signal is only here at Function block F2 and supplied to the PVE level comparison element, the resulting output signals the function block F3 to pass to the speakers L and R. In function block F3 takes place according to the partial function F3 processing of monaural input signal of the receiver side Data terminal A.

0040The combination of two functional blocks represents a high quality, yet affordable option is where the quality of the three-dimensional simulation with each vote regardless of the area of ​​operation of the headset is.

0041The change of the monaural signal by means of one of these Processing element also generates an auditory event, the at least components of the natural hearing reflects. Thus it is in application of the proposed headset possible to locate different sound sources, and in particular to suppress the perception of discards. This is because of the modeled natural hearing, in which the human being to suppress the perception of Discards has actually learned.

0042The exclusive use of individual functional blocks as Runtime elements and / or damping elements and / or HRTF processing elements already makes a special Sound sensation that's example sufficient if while communicating little background noise.

0043It should be noted that all the above-described Parts seen on its own and in any combination, especially the details shown in the drawing are claimed as essential to the invention. amendments thereof to those skilled in the art. Thus, the Place the aforementioned transit time or delay elements Means for sign inversion of the processed signals to step.