Voice control system with a microphone array

[0001] [Technical field to which the invention belongs] The present invention relates to a voice control system having a microphone array and particularly capable of being used to control a device belonging to a consumer electronic device. [0002] [Conventional technology] Voice control systems are used in many technical fields. In this case, the spoken word is usually first detected as a sound signal by one or more microphones and then fed to a speech recognition system. In this case, speech recognition is usually based on acoustic and speech models. Acoustic models utilize a large number of speech patterns and mathematical algorithms are used to indicate the words that best acoustically match the spoken word. The speech model is based on an analysis that uses a large number of document samples to ascertain the context and frequency in which a word is commonly used. Such a speech recognition system makes it possible to recognize not only individual words but also sentences fluently spoken with a high recognition rate. However, in the presence of non-negligible background noise, the recognition rate is greatly reduced. [0003] The resilience to the effects of such acoustic interference can be increased in various ways. Therefore, in a dictation system for a computer, the microphone provided on the headset frame is fixed just in front of the speaker's mouth. In these systems, a very constant signal, and in some circumstances a significant recognition rate, can only be achieved in the immediate vicinity of the mouth. Similarly, it is known to control a television receiver by speaking an operation command to a microphone integrated with a remote control device. However, again, the remote controller must be held just in front of the user's mouth. [0004] [Problems to be Solved by the Invention] It is an object of the present invention to provide a voice control system that enables sufficient safety against interference even in the case of voice input from a relatively long distance. [0005] [Means for solving problems] The above object is achieved by the system according to claim 1. The voice signal must be separated from the disturbing background signal to allow voice control from a relatively long distance . This can be done by separating into space using a microphone array containing two or more microphones. In this case, it is advantageous that the individual microphones in the microphone array are spatially dispersed over the largest possible distance. However, in individual consumer electronic devices, the distance between individual microphones is limited by the dimensions of the device, for example, in the case of television receivers, to 1 meter or less. [0006] In principle, the voice control system according to the present invention has a plurality of microphones that convert voice commands into electric signals, and a central voice recognition unit that converts the electric signals into operation commands. The generated signal is distributed among different interconnected devices so that it can be transmitted to the central speech recognition unit. [0007] In this case, it is particularly advantageous that the devices are connected via a bidirectional network, which is an IEEE1394 bus. [0008] The system is particularly advantageous when one or more microphones are integrated into consumer regenerative electronics, especially television receivers, and one or more additional microphones are integrated into external speakers. [0009] BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, the present invention will be described with reference to the accompanying drawings. [0010] FIG. 1 is a diagram schematically showing a system according to the present invention. Two external speakers LS1 and LS2 are connected to the television receiver TV. The internal speakers (not shown) of the television receiver, along with the external speakers, enable surround sound reproduction of multi-channel audio signals, such as the reproduction of corresponding digital audio signals conforming to the MPEG2 or AC3 standards. In this example, the external speakers are connected to the television receiver via an IEEE1394 bus, also known as FireWire, but may be connected to a suitable surround sound receiver. Using the IEEE1394 bus is advantageous because it enables high-speed data transmission and communication between different devices. Further, in the case of an active speaker, power can be supplied simultaneously via this bus. [0011] A microphone array is provided to detect the voice signal. The microphone array includes two microphones MTV1 and MTV2 integrated in the television receiver, and the microphones MLS1 and MLS2 are integrated in the speaker housing. These microphones convert the detected sound signal into an electric signal, which is amplified by an amplifier, converted into a digital signal by an AD converter, and supplied to a signal processing unit. In this example, the signal from the external speaker is similarly supplied to the signal processing unit in the television receiver via the IEEE1394 bus. The signal processing unit takes into account each user's position by different scaling or processing of the detected sound signal. In addition, the microphone signal can be corrected with respect to the sound signal output by the speaker. The signal processed in this way is then supplied to a speech recognition unit that converts the electrical signal into words. Finally, the instructions corresponding to these words are fed to the system manager to control the system. [0012] Scaling or processing of the detected sound signal by the signal processing unit requires that the spatial arrangement of the microphones be known. This is known by the manufacturer for microphones integrated into the television enclosure. On the other hand, in the case of a microphone arranged in a speaker, the position with respect to the television receiver must be determined. This is done by measuring and inputting values determined via the corresponding screen menus. However, measurement and configuration operations can also be performed by the test signal tone being reproduced by the speaker, detected by the microphone, and the position of the microphone being determined from various propagation delays. [0013] The two microphones integrated into the television receiver can be advantageously housed on the left and right sides of the housing of each device. However, the number of microphones and the number of additional microphones is not limited to two. Many combinations are conceivable in which the microphone is integrated. Thus, the microphone may be integrated into a video recorder, DVD player, or remote controller instead of or in addition to the television. It is also possible to incorporate the microphone into devices located in different rooms. [0014] Furthermore, device connections are not limited to bus systems. Therefore, for example, it is conceivable that the speaker of the embodiment of FIG. 1 is driven by a wireless signal. However, in this case, each speaker must have a wireless transmitter in addition to the wireless receiver. [0015] The present invention relates to consumer electronic devices such as television receivers, video recorders, DVD players, satellite receivers, television / video combinations, audio equipment, or similar complete audio systems for personal computers or home appliances. Can be used for voice-operated remote control of a wide range of devices belonging to. [Simple explanation of drawings] FIG. 1 is a diagram showing an arrangement for voice control using a microphone array according to the present invention. [Explanation of symbols] MLS1, MLS2 microphone LS1, LS2 external speaker TV television receiver MTV1, MTV2 microphone