JPH0691578B2 - Voice switch circuit - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a howling prevention voice switch circuit used in a two-way communication device such as a speakerphone or a teleconference system.

2. Description of the Related Art In recent years, speakerphones and teleconference systems have been remarkably popularized, and in these devices, a voice switch circuit is used to prevent howling.

In order to improve the performance of these devices, there is a strong demand for a voice switch circuit that does not malfunction.

A conventional voice switch circuit will be described below with reference to FIGS. 3 and 4 with reference to the drawings.

FIG. 3 shows a block diagram of a conventional voice switch circuit.

In FIG. 3, reference numeral 31 is an input terminal of a reception input signal, 32 is a speaker 33 which attenuates the reception input signal and serves as a reception output signal.
To the output attenuator, 33 is a speaker, 34 is a microphone, and 35 is a transmit attenuator that attenuates the transmit input signal output from the microphone 34 and outputs it as an output signal to the output terminal 36. The output terminal of the talk output signal, 37 is a reception amplitude calculator for rectifying and smoothing the reception input signal to obtain the reception amplitude, 38
Is a transmission amplitude calculation unit that rectifies and smoothes the transmission input signal to obtain the transmission amplitude, 39 is a division unit that divides the transmission amplitude by the reception amplitude to obtain the amplitude ratio, and 40 is this amplitude ratio and is preset. When the amplitude ratio is equal to or more than the threshold value, it is determined to be transmission, and when the amplitude ratio is less than the threshold value, it is determined to be reception. In the receiving state, the attenuation amount of the receiving attenuating unit 32 is set low, and the attenuation amount of the transmitting attenuating unit 35 is set high. Conversely, in the transmitting state, the attenuation amount of the receiving attenuating unit 32 is high and the receiving attenuating unit is high. The attenuation of 35 is set low.

The operation of the conventional voice switch circuit configured as described above will be described below.

FIG. 4 shows signal waveforms at various portions of the conventional example shown in FIG. 3, and the operation of FIG. 3 will be described with reference to this figure.

In FIG. 4, 41 is the waveform of the reception input signal at point A in FIG. 3, 42 is the waveform of the transmission input signal at point B in FIG. 3, 43 is the waveform of the reception amplitude at point C, and 44 is the waveform at point D. Waveform of transmission amplitude, 45 is E
The waveform of the amplitude ratio of the points, 46 is the threshold level inside the transmission / reception determination unit 40, 47 is the waveform of the determination signal at the F point, and 48 is the reception attenuation unit 32.
The reference numeral 49 represents the change in the attenuation amount of the transmission attenuation unit 35, and the reference numeral 49 represents the change in the attenuation amount of the transmission attenuation unit 35.

In a two-way call, the speaker on the terminal side of interest is called the near-end speaker, and the speaker on the other side is called the far-end speaker, and the terms will be used hereinafter.

This is the voice signal of the far-end speaker. Receiving input signal 41 in FIG.
Is rectified and smoothed by the reception amplitude calculator 37 in FIG. 3 to become the reception amplitude 43 in FIG. On the other hand, this reception input signal 41 is the third
It passes through the reception attenuator 32 in the figure to become a reception output signal, which is amplified by the speaker 33. The microphone 34 picks up the reverberant sound from the speaker 33 and the voice of the near-end speaker talking to the microphone 34, and outputs a transmission input signal 42. This transmission input signal 42 is rectified and smoothed by the transmission amplitude calculation unit 38, and becomes the transmission amplitude 44. On the other hand, the transmission input signal 42 passes through the transmission attenuation unit 35 to become a transmission output signal, which is output from the output terminal 36. The division unit 39 divides the transmission amplitude 43 by the reception amplitude 44 to obtain the amplitude ratio 45. The transmission / reception determination unit 40 compares the threshold value 46 with this amplitude ratio 45. When the amplitude ratio is equal to or higher than the threshold value, it is determined to be transmission.
To the transmission attenuation unit 35. The attenuation amounts of the two attenuating units are reciprocally controlled by the determination signal 47, and when one of the attenuation amounts is high, the other attenuation amount is always low.

That is, in the receiving state, the attenuation amount of the reception attenuation unit 32 is small, and the attenuation amount of the transmission attenuation unit 35 is large. On the contrary, in the transmission state, the attenuation amount of the reception attenuation unit 32 is large and the attenuation amount of the transmission attenuation unit 35 is small. As described above, the attenuation amount of the reception attenuation unit 32 is controlled as indicated by 48, and the attenuation amount of the transmission attenuation unit 35 is controlled as indicated by 49.

In this way, the amplitude of the reception input signal and the amplitude of the transmission input signal are compared, and the one with the larger amplitude is output without being attenuated and the one with the smaller amplitude is output after being attenuated. That is, of the near-end speaker and the far-end speaker, the speaker voice speaking in a louder voice is preferentially output. Therefore, it is possible to reduce the loop gain, and it is possible to talk in a large volume without howling.

Problems to be Solved by the Invention However, the above-described configuration has a drawback in that when the acoustic coupling of the echo path between the speaker 33 and the microphone 34 is too large, an erroneous operation called reception blocking occurs. When the acoustic coupling of the echo path is large and the received input signal is amplified by the speaker 33 without being attenuated, a high-level echo sound enters the microphone 34.

As a result, the value of the amplitude ratio becomes extremely large and exceeds the threshold value, and it is erroneously determined that the near-end speaker has started speaking even though the far-end speaker is speaking. Therefore, attenuation is inserted in the reception input signal. That is,
The far-end talker's voice is cut off, resulting in a very unnatural call.
To prevent this phenomenon, the threshold value should be set high in advance, but if it is set higher than necessary, this time the sensitivity to the transmission input signal, that is, the near-end speaker's voice decreases, and the head disconnection is the target. Then, the call quality deteriorates.

Since the conventional configuration does not have the function of accurately estimating the acoustic coupling magnitude of the echo path from the speaker 33 to the microphone 34, it is impossible to control the threshold value by following the acoustic coupling magnitude. Met.

In view of the above problems, the present invention can control the threshold value by following the size of the acoustic coupling of the echo path between the speaker and the microphone, even if the microphone, the position of the speaker, and the reverberation characteristic of the room change. Provided is a voice switch circuit capable of detecting a transmitted signal with high accuracy without causing reception blocking.

Means for Solving the Problems In order to achieve this object, the voice switch circuit of the present invention attenuates the reception input signal where the attenuation amount is small during reception and is controlled to be large during transmission and reception. A reception attenuator that outputs an output signal, a transmission attenuator that attenuates a transmission input signal that is controlled to have a large attenuation amount during reception and a small attenuation amount during transmission, and the reception input A receiver input amplitude calculator that determines the amplitude of the signal,
A reception output amplitude calculation unit that determines the amplitude of the reception output signal, a transmission input amplitude calculation unit that determines the amplitude of the transmission input signal, and the amplitude of the transmission input signal is divided by the amplitude of the reception input signal. A first division unit for obtaining a transmission / reception determination amplitude ratio,
A second divider for obtaining the amplitude ratio for acoustic coupling measurement by dividing the amplitude of the transmission input signal by the amplitude of the reception output signal, and the amplitude ratio for acoustic coupling measurement that fluctuates repetitively by repeating peak and dip. A peak hold unit that obtains the envelope of the peak with the upper end of the peak swept, and a dep hold unit that obtains the minimum peak value that is the minimum value of the envelope of this peak,
If a threshold value calculation unit that sets a threshold value from the minimum peak value and the threshold value and the transmission / reception determination amplitude ratio are compared, and the transmission / reception determination amplitude ratio is larger than the threshold value, It is constituted by a transmission / reception determining unit that determines that the transmission is a reception, if the transmission is small, and that the reception is a reception, and transmits the determination result to the reception attenuation unit and the transmission attenuation unit.

Action With this configuration, the magnitude of acoustic coupling in the echo path from the speaker to the microphone can be measured at all times, and the threshold value can be automatically set to match the measured value. For this reason, transmission / reception determination can be performed with high sensitivity and without erroneous detection, and two-way communication without howling or disconnecting the talk can be performed.

Embodiment An embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, reference numeral 1 is an input terminal of a reception input signal, 2 is a reception attenuating unit for attenuating the reception input signal to obtain a reception output signal,
3 is a speaker, 4 is a microphone, 5 is a microphone 4
Of the transmission input signal, which is the output of the transmission input signal, is attenuated as a transmission output signal, 6 is an output terminal of the transmission output signal, and 7 is rectified and smoothed to the reception input signal to obtain the amplitude of the reception input signal. Receiving input amplitude calculating unit, 8 is a receiving output amplitude calculating unit for rectifying and smoothing the receiving output signal to obtain the amplitude of the receiving output signal, and 9 is a transmitting voice for rectifying and smoothing the transmitting input signal to obtain the amplitude of the transmitting input signal. An input amplitude calculator, 10 is a first divider that divides the amplitude of the transmission input signal by the amplitude of the reception input signal to obtain a transmission / reception determination amplitude ratio, and 11 is the amplitude of the transmission output signal. The second division section for obtaining the amplitude ratio for acoustic coupling measurement by dividing by 12, 12 is a peak for obtaining the envelope curve of the peak in which the upper end of the peak of the amplitude ratio for acoustic coupling measurement that fluctuates repetitively by repeating the peak dip is obtained. Hold part, 13 is the minimum value of the envelope of this peak. A dephold unit for obtaining the minimum peak value, 14 is a threshold value calculation unit that sets a threshold value from the minimum amplitude ratio and the minimum peak value, and 15 compares this threshold value with the transmission / reception determination amplitude ratio. Then, when the transmission / reception determination amplitude ratio is larger than the threshold value, the transmission / reception determination unit determines transmission, and when the transmission / reception determination amplitude ratio is smaller than the threshold value, the transmission / reception determination unit 15 determines that the reception signal of the reception attenuation unit 2 The attenuation amount is low, the attenuation amount of the transmission attenuation unit 5 is set high, and conversely in the transmission state, the attenuation amount of the reception attenuation unit 2 is high and the attenuation amount of the transmission attenuation unit 5 is set low.

FIG. 2 shows signal waveforms at various portions of the embodiment shown in FIG. 1, and the operation of the embodiment shown in FIG. 1 will be described with reference to this drawing.

In FIG. 2, 21 is a waveform of the reception input amplitude at point A in FIG. 1, 22 is a waveform of the reception output amplitude at point B of FIG. 1, 23 is a waveform of the transmission input amplitude at point C, and 24 is D. Waveform of transmission / reception determination amplitude ratio of point, 25 is waveform of amplitude ratio for acoustic coupling measurement of point E, 251 is waveform of envelope of peak of point F, 252 is waveform of minimum peak value of point G, 26 is H The waveform of the threshold value of the point, 27 is the waveform of the determination signal of the point I, 28 is the state of the change in the attenuation amount of the reception attenuation unit 2,
Reference numeral 29 indicates how the attenuation amount of the transmission attenuation unit 5 changes.

The reception input signal, which is the voice signal of the far-end speaker, is rectified and smoothed by the reception input amplitude calculation unit 7 in FIG. 1 and becomes the reception input amplitude 21 in FIG. On the other hand, the received voice input signal passes through the received voice attenuator 2 in FIG. 1 to become a received voice output signal, which is amplified by the speaker 3. This reception output signal is rectified and smoothed by the reception output amplitude calculation section 8 and becomes the reception output amplitude 22 shown in FIG. The microphone 4 receives the reverberant sound from the speaker 3 and the microphone 4
It collects the voice of the near-end speaker who is speaking to and outputs the transmission input signal. The transmission input signal is rectified and smoothed by the transmission input amplitude calculation unit 9 to have the transmission input amplitude 23. On the other hand, this transmission input signal passes through the transmission attenuation unit 5 to become a transmission output signal, which is output from the output terminal 6. First division
In 10, the transmission input amplitude 23 is divided by the reception input amplitude 22 to obtain the transmission / reception determination amplitude ratio 25. The second divider 9 divides the transmission input amplitude 24 by the reception output amplitude 23 to obtain the acoustic coupling measurement amplitude ratio 25.

At times T _{1 to} T ₂ shown in Fig. 2 in which only the far-end speaker is speaking and in the receiving state, the acoustic coupling measurement amplitude ratio 25 is the magnitude of acoustic coupling in the echo path composed of the speaker and the microphone. Indicates. At time T ₃ through T ₄ near end speaker is Hanashidashi, because only the sending input amplitude 23 is increased, the amplitude ratio for acoustic coupling measurement
25 is a very large value. Therefore, in order to determine the magnitude of acoustic coupling in the echo path, this acoustic coupling measurement amplitude ratio 25
It can be seen that the minimum value of However,
As can be seen from the acoustic coupling measurement amplitude ratio 25 at times T _{1 to} T _2, the acoustic coupling measurement amplitude ratio 25 slightly changes. The degree of this variation greatly depends on the characteristics of the echo path. Therefore, the minimum value of the acoustic coupling measurement amplitude ratio 25 cannot be set as the acoustic coupling magnitude of the echo path. In the present invention, the acoustic coupling strength of the echo path is accurately calculated by the method described below.

First, the peak hold unit 12 obtains a peak envelope 251 in which the upper end of the peak of the acoustic coupling measurement amplitude ratio 25 is swollen.
Next, the dep hold unit 13 obtains the minimum peak value 252 which is the minimum value of the envelope 251 of this peak. Looking at this minimum peak value 252 in detail, it can be seen that the minimum peak value 252 represents the maximum value of the acoustic coupling strength of the echo path. Therefore, the threshold value calculation unit 14 calculates the threshold value 26 from the minimum peak value 252 by the following equation.

Threshold = 2 * minimum peak value (1) The transmission / reception determination unit 15 compares the threshold value 26 thus obtained with the transmission / reception determination amplitude ratio 24 to determine the transmission / reception determination amplitude ratio. When the value is equal to or more than the value, it is determined that the speech is transmitted, and when it is less than the value, it is determined that the speech is received, and the determination signal 27 is sent to the reception attenuation section 2 and the transmission attenuation section 5. The attenuation amounts of the two attenuating parts are controlled reciprocally, and when the attenuation amount of one increases, the attenuation amount of the other always decreases. That is, in the receiving state, the attenuation amount of the reception attenuation unit 2 is small and the attenuation amount of the transmission attenuation unit 5 is large. On the contrary, in the transmission state, the attenuation amount of the reception attenuation unit 2 is large and the attenuation amount of the transmission attenuation unit 5 is small. As described above, the reception attenuation unit 2
As shown in 28, the transmission attenuation unit 5 has 29
It is controlled as shown in.

In this way, a threshold value corresponding to the acoustic coupling strength of the echo path can be automatically set from the value of the acoustic coupling measurement amplitude ratio. Therefore, it is possible to always perform the voice switch operation with high sensitivity and less false detection.

EFFECTS OF THE INVENTION The present invention has a peak hold section for obtaining an envelope curve of a peak having the upper end of a peak of an amplitude ratio for acoustic coupling measurement, and a dep hold section for obtaining a minimum peak value which is the minimum value of the envelope curve of the peak. A threshold value calculating unit that sets a threshold value from this minimum peak value constitutes a threshold value setting means, and the echo path is calculated from the amplitude ratio for acoustic coupling measurement that fluctuates repetitively by repeating peak-dep. It is possible to accurately obtain the acoustic coupling strength of and to automatically set a threshold value corresponding to the acoustic coupling strength.

If the acoustic coupling strength of the echo path is large, the threshold value is high, and if the acoustic coupling strength is small, the threshold value is low. realizable.

[Brief description of drawings]

FIG. 1 is a block diagram of a voice switch circuit according to an embodiment of the present invention, FIG. 2 is a waveform diagram of signals in each part of FIG. 1, FIG. 3 is a block diagram of a conventional voice switch circuit, and FIG. FIG. 4 is a waveform chart of signals in each part of FIG. 3. 2 ... Reception attenuation unit, 5 ... Transmission attenuation unit, 7 ... Reception input amplitude calculation unit, 8 ... Reception output amplitude calculation unit, 9 ... Transmission input amplitude calculation unit, 10 ... First division Section, 11 ... Second division section, 12 ... Peak hold section, 13 ... Dep hold section, 14 ... Threshold calculation section, 15 ... Transmission / reception determination section.

Claims (1)

[Claims] 1. A reception attenuating unit for attenuating a reception input signal to produce a reception output signal, where the attenuation amount is small during reception and the attenuation amount is controlled large during transmission, and the attenuation amount is large during transmission and attenuation during transmission. A transmission attenuation unit that attenuates a transmission input signal whose amount is controlled to be a small amount as a transmission output signal, a reception input amplitude calculation unit that obtains the amplitude of the reception input signal, and an amplitude of the reception output signal. A desired reception output amplitude calculation unit, a transmission input amplitude calculation unit that determines the amplitude of the transmission input signal, and an amplitude ratio for transmission / reception determination by dividing the amplitude of the transmission input signal by the amplitude of the reception input signal. A first divider, a second divider that divides the amplitude of the transmission input signal by the amplitude of the reception output signal to obtain an amplitude ratio for acoustic coupling measurement, and a peak-deep repetition that fluctuates drastically. Amplitude for acoustic coupling measurement The peak hold part that finds the envelope of the peak with the upper end of the peak of, and the dep hold part that finds the minimum peak value that is the minimum value of the envelope of this peak, and the threshold value is set from this minimum peak value. A threshold value calculation unit is compared with this threshold value and the transmission / reception determination amplitude ratio, and when the transmission / reception determination amplitude ratio is larger than the threshold value, it is determined to be transmission. An audio switch circuit comprising: a transmission / reception determination unit for transmitting a result to the reception attenuation unit and the transmission attenuation unit.