JP2529361B2 - Voice switch - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a howling-preventing voice switch used in a public telephone, a teleconference system, or the like.

2. Description of the Related Art In recent years, two-way communication devices such as a loudspeaker telephone and a teleconference system have been remarkably spread, and these devices use a voice switch for preventing howling.

Hereinafter, a conventional voice switch will be described with reference to the drawings.

FIG. 5 shows a block diagram of a conventional voice switch.

In FIG. 5, 501 is an input terminal for receiving a reception input signal from the line, 502 is a reception attenuator that attenuates the reception input signal and outputs it as a reception output signal to a speaker amplifier 503, 503 is a speaker amplifier, and 504 is a speaker. , 505 is a microphone, 506 is a microphone amplifier, 507 is a microphone amplifier 506
Attenuator that attenuates the transmission input signal that is the output of the transmission output signal to the output terminal 508 as a transmission output signal, 508 is an output terminal that outputs the transmission output signal to the line, and 509 is a receiver input signal amplitude device. A reception amplitude calculator that obtains a reception input amplitude value that is
Reference numeral 501 is a transmission amplitude calculation unit that obtains a transmission input amplitude value that is the amplitude value of the transmission input signal, and 511 is a comparison between the reception input amplitude value and the transmission input amplitude value, and the reception input amplitude value is greater than or equal to the transmission input amplitude value. When the signal is in the receiving state, in the following cases the transmission / reception determination unit that determines the transmission state, and 512 is the attenuation amount of the reception attenuator 502 is small when the determination signal of the transmission / reception determination unit 511 is the reception state. This is an attenuation amount control unit that sets a large amount and, conversely, sets a small amount of attenuation of the reception attenuator 502 and a large amount of attenuation of the transmission attenuator 507 in the transmitting state.

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

FIGS. 6 (a) to 6 (i) show signal waveforms at respective portions of the conventional example of FIG. 5, and the operation of FIG. 5 will be described with reference to this figure.

6, 601 is the waveform of the reception input signal at point A in FIG. 5, 602 is the waveform of the reception output signal at point B in FIG. 5, 603 is the waveform of the transmission input signal at point C, and 604 is the point D. Waveform of the transmission output signal of 605, 605 is the reception input amplitude value of the point E, 606 is the transmission input amplitude value of the point F, 607 is the waveform of the transmission / reception determination signal of the point G, and 608 is the attenuation amount of the reception attenuator 502. Setting state, 609 is transmission attenuator 507
3 shows the setting state of the attenuation amount of.

In a two-way call, the terminal side is called the near end, the partner terminal side is called the far end, the near-end speaker is called the near-end speaker, and the far-end speaker is called the far-end speaker. Hereinafter, description will be made using this term.

The reception input signal 601 of FIG. 6 which is the voice signal of the far-end speaker.
Becomes a reception output signal 602 through the reception attenuator 502 of FIG. The reception output signal 602 is amplified by the speaker amplifier 503 and is amplified by the speaker 504 on the near end side. Microphone 5
05 picks up the reverberant sound from this speaker 504 and the voice of the near-end speaker talking to the microphone 505. The output is amplified by the microphone amplifier 506 and becomes the transmission input signal 603. The transmission input signal 603 passes through the transmission attenuator 507 to become the transmission output signal 604, which is output from the output terminal 508 to the far end. The reception amplitude calculator 509 calculates the amplitude value of the reception input signal 601, and obtains the reception input amplitude value 605. Transmitting input amplitude calculator 5
At 10, the amplitude value of the transmission input signal 603 is calculated to obtain the transmission input amplitude value 606. In the transmission / reception determination unit 511, the reception input amplitude value 605
When the transmission input amplitude value is 606 or more, it is determined that the transmission is received, and when the transmission input amplitude value is less than 606, transmission is determined, and the transmission / reception determination signal 607 is sent to the attenuation amount control unit 512.
The attenuation amount control unit 512 controls the attenuation amount of the reception attenuator 502 to be small when the transmission / reception determination signal 607 is in the reception state, and controls the attenuation amount of the transmission attenuator 507 to be small when the transmission / reception determination signal 607 is in the reception state.
The attenuation amounts of these two attenuators are reciprocally controlled so that the attenuation amount of the other one becomes smaller when the attenuation amount of one is larger, and as a result, the reception attenuator 502 becomes the attenuation amount shown by 608. The transmission attenuator 507 is controlled to the attenuation amount indicated by 609.

With the above configuration, it is determined which of the near-end speaker and the far-end speaker is speaking loudly, and by reciprocally controlling the attenuation amounts of the receiving system and the transmitting system based on the determination result, Of the near-end speaker and the far-end speaker, the speaker who is speaking louder is given priority to the loud voice, and the volume is increased without increasing the loop gain, that is, without howling. It enables voice calls over the phone.

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention However, in the above-described configuration, the speaker voice of the one with a smaller voice is rapidly attenuated during the double talk in which the near-end speaker and the far-end speaker simultaneously speak, and therefore, the acronym cutting is performed. There was a problem that the feeling became large and the naturalness and intelligibility of the conversation decreased.

In order to solve this problem, the two attenuators are not controlled by switching them to two levels of large and small, but the attenuating amounts of the two attenuators are controlled steplessly according to the amplitude levels of both voices. There is a need. The conventional configuration is a system in which two states, that is, transmission and reception, are determined, and the amount of attenuation is switched to two levels depending on the determination result, and such control is impossible.

In view of the above problems, the present invention provides a voice switch that allows a natural and well-understood call even during double talk.

Means for Solving the Problems In order to achieve this object, a voice switch of the present invention includes a reception attenuator that attenuates a reception input signal and outputs it as a reception output signal, and a transmission attenuator that attenuates a transmission input signal. A transmitter attenuator that outputs as an output signal, a receiver amplitude calculator that determines the receiver input amplitude value that is the amplitude value of the receiver input signal, and a transmitter that determines the transmitter input amplitude value that is the amplitude value of the transmitter input signal. A speech vibration calculation unit, a transmission / reception amplitude ratio calculation unit that obtains a transmission / reception amplitude ratio that is a ratio of the reception input amplitude value and the transmission input amplitude value, a set value of the attenuation amount of the two attenuators, and the transmission / reception amplitude ratio. From the value of, the voice amplitude ratio estimation unit that estimates the voice amplitude ratio, which is the ratio between the amplitude value of the far-end speaker voice and the amplitude value of the near-end speaker voice, and the estimation estimated by the voice amplitude ratio estimation unit. Calculate the attenuation that should be newly set from the value of the voice amplitude ratio, and Two
It is composed of an attenuation amount control unit that controls the attenuation amount of one attenuator.

With this configuration, the voice amplitude ratio, which is the amplitude ratio between the near-end speaker voice and the far-end speaker voice, can be estimated, and the attenuation amounts of the two attenuators are controlled according to the estimated voice amplitude ratio,
The attenuation amounts of the two attenuators can be set to be substantially equal when the amplitude values of the near-end talker voice and the far-end talker voice are almost equal to each other as in the case of double talk, that is, when the voice amplitude ratio is close to 1. Therefore, it is possible to overcome the conventional problem that the voice of the speaker with a lower voice is excessively attenuated during the double talk, and to provide a voice switch with less feeling of initial word ending and excellent conversation naturalness and intelligibility. .

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

In FIG. 1, 101 is an input terminal for receiving a reception input signal from a line, 102 is a reception attenuator that attenuates the reception input signal and outputs it as a reception output signal to a speaker amplifier 103, 103 is a speaker amplifier, and 104 is Speaker, 105 microphone, 106 microphone amplifier, 107 microphone amplifier 10
A transmission attenuator that attenuates the transmission input signal that is the output of 6 and outputs it to the output terminal 108 as a transmission output signal, 108 is an output terminal that outputs the transmission output signal to the line, and 109 is the reception input signal Of the receiving input amplitude value which is the amplitude value of the receiving input amplitude calculation unit, 110 is a transmitting amplitude calculation unit which determines the transmitting input amplitude value which is the amplitude value of the transmitting input signal, 111 is the receiving input amplitude value and the A transmission / reception amplitude ratio calculation unit for obtaining a transmission / reception amplitude ratio, which is a ratio of the transmission input amplitude value, and 112, from the set value of the attenuation amount of the two attenuators and the transmission / reception amplitude ratio value, the amplitude of the far-end speaker's voice. Value, and a voice amplitude ratio estimation unit for estimating the amplitude value of the near-end speaker's voice, 113 calculates an attenuation amount to be newly set from the estimated voice amplitude ratio estimated by the voice amplitude ratio estimation unit, and 2
It is an attenuation amount control unit that controls the attenuation amounts of the two attenuators.

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

Prior to the description of the operation of the embodiment shown in FIG. 1, first, the estimation principle of the voice amplitude ratio and the calculation principle of the attenuation amount in this embodiment will be briefly described using the analytical model shown in FIG.

In FIG. 2, the far-end speaker is speaking at the level of Vr and the near-end speaker is at the level of Vs. At this time, the receiving input signal is Ri, the receiving output signal is Ro, the transmitting input signal is Si, The transmission output signal is So
, The gain of the reception attenuator is set to Gr, and the gain of the transmission attenuator is set to Gs. In the figure, α is the reception attenuator 102
From the output to the input of the transmission attenuator 107, β represents the gain from the output of the transmission attenuator 107 to the input of the reception attenuator 102. The following relational expression holds in this model.

Ri ＝ Vr ＋ β ・ So Ro ＝ Gr ・ Ri Si ＝ Vs ＋ α ・ Ro (1) So ＝ Gs ・ Si The transmission / reception amplitude ratio: The theoretical value of L can be expressed by equation (2). Therefore, when the equation (2) is transformed into an equation having the normalized amplitude ratio: N obtained by dividing the transmission / reception amplitude ratio: L by α · Gr, the equation (3) is obtained. In this equation, the voice amplitude ratio (Vs / Vr), which is the amplitude ratio of the near-end speaker's voice and the far-end speaker's voice, is represented by K, and the transmission input amplitude value: Si and the reception input amplitude value: Ri A ratio of transmission / reception amplitude: (Si / Ri) is represented by L, and a normalized amplitude ratio is represented by N.

FIG. 3 illustrates the equation (3).

Curves 31 to 39 in FIG. 3 are plots of the N value and the K value on the horizontal axis in the state of 9 kinds of Gr values with α and β as constants. At this time, Gr and Gs are reciprocally controlled as shown in equation (4).

Gr = Hm / Gs (4) In the equation (4), Hm is a constant value that is sufficiently smaller than 1.

A curve 30 in FIG. 3 shows a locus drawn by N when Gr and Gs are controlled by the method according to the present invention, and FIG. 4 shows the setting method of Gr and Gs.

L in the equation (3) is a value obtained by the amplitude ratio calculation unit. The voice amplitude ratio estimation unit first divides this L by Gr and α to obtain N, and then uses FIG. 3 to estimate K from Gr, Gs, and N. The attenuation control unit determines Gr and Gs from FIG. 4 and K.

Next, the operation of the voice switch of the present invention using FIGS. 3 and 4 as described above will be described in more detail.

First, when the Gr value is set to Gr1 and the Gs value is set to Gs1, N1
Suppose that an N value of That is, FIG. 3 (Gr1,
Gs1, K1, N1) is assumed to be in the state of point A. At this time, the voice amplitude ratio estimation unit estimates a K value K1 from Gr1, Gs1, N1 and FIG. From Fig. 3, K1 in this state
It can be seen that the following estimated value is obtained. Next, the attenuation control unit determines the Gr value Gr2 and the Gs value Gs2 from K1 and FIG. 4, and updates Gr and Gs. As a result, it can be seen that the curve 30 shown in FIG. 3 is stable at point B indicated by (Gr2, Gs2, K1, N2). In this FIG. 3, it is assumed that only the far-end speaker is speaking as K1, and that Gr >> Gs is set when only the far-end speaker is speaking.

Next, interrupt the near-end speaker, where the K value is K2,
That is, it is assumed that there is a call interruption. This time,
Since the Gr value is set to Gr2 and the Gs value is set to the Gs2 value, immediately after the interruption, the transition to the C point (Gr2, Gs2, K2, N3) occurs and the N value N3 is obtained. The speech amplitude ratio estimation unit uses Gr2, Gs2, N3, and the third
Estimate the K value of K2, which is
Then, determine the values Gs3 and Gs3 from Fig. 4 and correct Gr and Gs. As a result, it stabilizes at point D (Gr3, Gs3, K2, N4) in FIG. It can be seen that the double talk state is assumed for K2, and that Gr ≈ Gs is set.

Next, when only the near-end speaker whose K value is K3 is speaking, immediately after the transition, the point E is (Gr3, Gs3, K3, N5) in FIG. In the attenuation estimation section, K3 is calculated from the values of Gr3, Gr3 and N5.
Estimate and correct Gr and Gs to the values Gr4 and Gs4. As a result, it becomes stable at point F (Gr4, Gs4, K3, N6) in FIG.
Therefore, if only the near-end speaker is speaking, Gr <<
You can see that it is set to Gs.

The above is the basic operation of the voice switch of the embodiment shown in FIG.

As described above, according to the present embodiment, by providing the transmission / reception amplitude ratio calculation unit, the voice amplitude ratio estimation unit, and the attenuation amount control unit,
It is possible to estimate the voice amplitude ratio of the far-end speaker's voice and the near-end speaker's voice, and to set the optimum amount of attenuation corresponding to this voice amplitude ratio, so that the voice amplitude ratio is close to 1 as in double talk. In this case, the attenuations of the two attenuators are set to the same level, and it is possible to completely overcome the conventional problem that the speaker's voice with a smaller voice is excessively attenuated during double talk, resulting in a sense of initial apical discontinuity. It is possible to realize a few excellent voice switches. Furthermore, since the transmission and reception are smoothly switched, the naturalness of conversation can be dramatically improved.

EFFECTS OF THE INVENTION The present invention provides a receiver attenuator that attenuates a receiver input signal and outputs it as a receiver output signal, a transmitter attenuator that attenuates a transmitter input signal and outputs it as a transmitter output signal, and the receiver input signal. A reception input amplitude value which is an amplitude value of the transmission input signal, a transmission amplitude calculation section which determines a transmission input amplitude value which is an amplitude value of the transmission input signal, the reception input amplitude value and the transmission signal The amplitude value of the far-end speaker's voice and the amplitude value of the far-end speaker are approximated from the transmission / reception amplitude ratio calculation unit for obtaining the transmission / reception amplitude ratio, which is the ratio of the input amplitude values, and the set value of the attenuation amount of the two attenuators and the transmission / reception amplitude ratio value. A voice amplitude ratio estimation unit that estimates a voice amplitude ratio that is a ratio with the amplitude value of the end-user voice, and calculates an attenuation amount to be newly set from the value of the estimated voice amplitude ratio estimated by the voice amplitude ratio estimation unit. However, an attenuation amount control unit for controlling the attenuation amounts of the two attenuators should be provided. With, it becomes possible to estimate the amplitude ratio between the far-end speaker's voice and the near-end speaker's voice, and set the optimum amount of attenuation in proportion to this voice amplitude ratio. It is possible to realize a voice switch.

[Brief description of drawings]

FIG. 1 is a block diagram of a voice switch according to an embodiment of the present invention, FIG. 2 is a model diagram for explaining an operation, FIG. 3 is an operation explanatory diagram for estimating a voice amplitude ratio, and FIG. 4 is for determining an attenuation amount. FIG. 5 is a block diagram of a conventional voice switch, and FIG. 6 is a waveform diagram of signals in each part of FIG. 102 ... Reception attenuator, 107 ... Transmission attenuator, 109 ... Reception input amplitude calculation unit, 110 ... Transmission input amplitude calculation unit, 111 ...
Transmission / reception amplitude ratio calculation unit, 112 ... Voice amplitude ratio estimation unit, 113 ...
Attenuation control unit.

Claims (1)

(57) [Claims] 1. A reception attenuator that attenuates a reception input signal and outputs it as a reception output signal, a transmission attenuator that attenuates a transmission input signal and outputs it as a transmission output signal, and an amplitude of the reception input signal. A reception input amplitude value which is a value, a transmission amplitude calculation section which determines a transmission input amplitude value which is an amplitude value of the transmission input signal, the reception input amplitude value and the transmission input amplitude The amplitude value of the far-end talker's voice and the near-end speech are calculated from the transmission-reception amplitude ratio calculation unit for obtaining the transmission-reception amplitude ratio, which is the ratio of the values, and the set value of the attenuation amount of the two attenuators and the value of the transmission-reception amplitude ratio. A voice amplitude ratio estimation unit that estimates a voice amplitude ratio that is a ratio to the amplitude value of the human voice, and calculates an attenuation amount to be newly set from the value of the estimated voice amplitude ratio estimated by the voice amplitude ratio estimation unit, and It is characterized by including an attenuation amount control unit for controlling the attenuation amounts of the two attenuators. Voice switch to.