JPH11154998A - Speech function equipment and its speaker voice extracting and transmitting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the size and weight of speech function equipment by registering the voice print of a speaker in advance prior to a speech, then, comparing it with the voice print of the speaker registered from a signal sound- collected from a transmitter at the time of speaking to extract and send the voice of the speaker and to unnecessitate a microphone for collecting an ambient noise and its circuit. SOLUTION: At the time of registering a voice print of a speaker, a switch 22 is thrown down to the side of a voice print analytic circuit 24 and a microphone collects the voice of the speaker to input the signal to the circuit 24 which analyzes the voice print of the speaker to store it in a voice print memory 26. Then, at the time of transmitting, the microphone collects the voice of the speaker and ambient noise to input the signal to a comparing filter 23. The comparing filter 23 inputs the voice print information of the speaker from the memory 26 at any time and extracts only the voice of the speaker from the signal inputted from the microphone from this information. The extracted voice signal is modulated by a transmission circuit and sent to a base station through an antenna.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a communication function device and a method for extracting and transmitting a speaker's voice, and more particularly to a communication function device for extracting a speaker's voice by removing ambient noise and a method for extracting and transmitting a speaker's voice. About the method.

[0002]

2. Description of the Related Art A conventional telephone function device of this type will be described with reference to the drawings.

FIG. 5 is a block diagram showing an example of a conventional telephone.

In FIG. 5, this conventional example is disclosed in
No. 22392, the first microphone 501 is a microphone that collects a speaker's voice, and the second microphone 502 is a microphone that collects ambient noise.

[0005] The ambient noise collected by the second microphone 502 passes through a low-frequency amplifier 321 and an A / D converter 322 and is input to an impulse response estimation circuit 505.

[0006] The impulse response estimation circuit 134 is a circuit for estimating an impulse response between the first microphone 501 and the second microphone 502 and generating pseudo noise.

[0007] The pseudo noise generated by the impulse response estimation circuit 505 passes through a D / A converter 323 and is added to the adder 5.
At 08, the signal collected by the first microphone 501 is subtracted. Thereby, ambient noise can be removed from the signal collected by the first microphone 501.

[0008]

The telephone as the conventional telephone function device has the following problems.

The first problem is that the telephone cannot be downsized. There is a trend toward smaller and lighter portable telephones, and the inability to reduce the size reduces the competitiveness of the portable telephone.

[0010] The reason is that two microphones are required.

A second problem is that ambient noise cannot be completely removed.

The reason is that if the second microphone (microphone collecting ambient noise) is at the same position as the first microphone (microphone collecting voice of a speaker), the noise is the same, Although it is possible to completely remove ambient noise, it cannot be mounted at the same position.

[0013] This is because if the position shifts, a difference occurs in the noise, and this difference becomes noise and is transmitted.

[0014]

SUMMARY OF THE INVENTION A call function device according to the present invention registers a voiceprint of a speaker in advance prior to a call, and then uses the voiceprint of the registered speaker from a signal collected from a transmitter during a call. Speaker voice extraction and transmission means for extracting and transmitting the voice of the speaker by comparison with the above.

In the method for extracting and transmitting a speaker's voice of the communication function device according to the present invention, a voiceprint of the communication is registered in advance before the communication.
At the time of a telephone call, the voice of the registered speaker is compared with the registered voiceprint from the signal collected from the transmitter, and the voice of the speaker is extracted and transmitted by this comparison.

[0016] The call function device of the present invention, prior to a call,
A voiceprint registration unit for registering a speaker's voiceprint in advance is provided.

According to another aspect of the present invention, there is provided a speaker function for making a speaker's voice talk by comparing a speaker or a voiceprint registered in advance with a signal collected from a transmitter when the speaker talks. And a pass filter means.

The passing function apparatus according to the present invention includes a switching unit for switching between voiceprint registration and a telephone call, and when the switching unit is switched to the voiceprint registration side, the voiceprint of the speaker from the speaker's voice is transmitted from the transmitter. , A voiceprint storage unit for storing the voiceprint of the speaker analyzed by the voiceprint analysis unit, and a signal collected from the transmitter when the switching unit is switched to the call side. A voice voice filter unit for extracting the voice of the speaker by comparing with the voice print of the speaker from the voice print storage unit.

According to the speaker voice extraction / call method of the call function device of the present invention, prior to a call, a switching section for switching between voiceprint registration and call is set on the voiceprint registration side, and a speaker from a transmitter is transmitted. The voiceprint of the speaker is analyzed from the voice by a voiceprint analysis unit, and the voiceprint of the speaker is stored in a voiceprint storage unit based on the result of the analysis. Then, the switching unit is set to the communication side, and the transmission The signal collected from the device and the speaker's voiceprint from the voiceprint storage unit are compared by the speaker's voice communication filter unit, and as a result, the speaker's voice is extracted and transmitted.

[0020]

Next, the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG. 2 is a block diagram showing a voice extracting circuit in the present embodiment.

In FIG. 1, a radio telephone according to the present embodiment includes a speech extraction circuit 12 for removing ambient noise from a speaker's voice from a signal collected from a microphone 11, and a speaker's voice from which the ambient noise has been removed. And a transmission circuit (TX) 13 for transmitting the signal to a base station (not shown) via an antenna 14.

Referring to FIG. 2, a voice extracting circuit 12 according to the present embodiment includes a voiceprint analyzing circuit 24 for analyzing a voiceprint of a speaker.
A voiceprint memory 26 that stores the voiceprint analyzed by the voiceprint analysis circuit 24, a switch 22 that switches between analyzing and storing the voiceprint of the speaker, and switching to a call state. And a comparison filter 23 for extracting the input signal by comparing it with the voiceprint.

Next, the operation of this embodiment and a method of extracting and transmitting a speaker's voice will be described with reference to FIGS.

First, the voiceprint of the speaker is registered. First, the switch 22 is turned to the voiceprint analysis circuit 24 side. A microphone 11 collects a speaker's voice and outputs the signal to a voice extraction circuit 1
2 is input to the voiceprint analysis circuit 24. Voiceprint analysis circuit 24
Analyzes the speaker's voiceprint and stores the result in the voiceprint memory 26. Now, the voiceprint registration of the speaker is completed, and the switch 2
2 is moved to the comparison filter 23 side for a call. This operation is not performed until the speaker changes.

Next, the operation during transmission will be described.
The microphone 11 collects the speaker's voice and ambient noise,
The signal is input to the comparison filter 23 of the audio extraction circuit 12. The comparison filter 23 reads from the voiceprint memory 26 as needed.
The voiceprint information of the speaker is input, and only the voice of the speaker is extracted from the signal input from the microphone 11 based on this information.

The extracted audio signal is transmitted to a transmission circuit (TX)
The signal is modulated by the antenna 13 and transmitted from the antenna 14 to the base station.

FIG. 3 is a block diagram showing an analog transmission-compatible radio telephone according to a first embodiment of the present invention.

Referring to FIG. 3, a radio telephone compatible with analog transmission according to the first embodiment includes a console unit (C) including dial keys, various mode setting switches, and a display unit.
U) 311 and a low-frequency amplifier 321 for amplifying a signal in the audio band from the microphone 11 to a predetermined level.
An A / D converter (A / D) 322 for converting the amplified signal into a digital signal; storing the voiceprint of the speaker; removing ambient noise from the input signal; and extracting the speaker's voice by the voiceprint 2 (however, only the switch 22 is in the CU 311) and a D / A converter (D / A) for converting the output of the voice extraction circuit 12 into an analog signal.
H.323, a synthesizer (SYN) 313 for generating a local transmission signal for modulation and demodulation, and FM modulation of a speaker voice signal from the D / A 323 with the local transmission signal, followed by frequency modulation with a high frequency signal and a duplexer (DUP) ) A transmission circuit (TX) 13 that transmits from the antenna 14 after passing through 314 and a radio signal that is received by the antenna 14 and passes through the duplexer (DUP) 314 are mixed by a local transmission signal from the SNY 313 and converted into an intermediate frequency. A receiving circuit (RX) 316 for demodulating the received voice signal and a speaker 31 for amplifying the demodulated received voice signal to a predetermined level;
8 and a control circuit (CONT) 312 for controlling the entire radio telephone.

Next, the operation of the first embodiment and a method of extracting and transmitting a speaker's voice will be described with reference to FIGS.

First, the speaker enters the console unit (C
U) The mode switch in 311 sets the voiceprint registration mode, and the signal switches the voice extraction circuit 12 to the voiceprint registration mode. The speaker inputs his / her own voice from the microphone 11, registers the voice in the voice extraction circuit 12, and thereafter inputs the CU3.
The mode switch in 11 is set to the call mode. Voiceprint registration does not execute this mode until the speaker changes.

Next, during a call, the transmitted voice signal collected by the microphone 11 as a transmitter is transmitted to the low-frequency amplifier 322.
The signal is amplified by the A / D converter (A / D) 322 and is converted into a digital signal. The voice extraction circuit 12 removes ambient noise from the transmitted voice signal by using the registered voiceprint, extracts only the voice of the speaker, and inputs the voice signal of the speaker to a D / A converter (D / A) 323. Return to analog signal.

In the transmission circuit (TX) 13, the transmission local transmission signal output from the frequency synthesizer (STN) 313 is frequency-modulated to a high-frequency signal after being FM-modulated by the transmission voice signal, and furthermore to a predetermined transmission power. Power is amplified. The high-frequency signal is transmitted to a duplexer (DUP)
After passing through 314, it is transmitted from the antenna 14 to the base station.

On the other hand, a radio signal arriving from a base station is received by an antenna 14 and then input to a receiving circuit (RX) 316 via a duplexer (DUP) 314. In the receiving circuit (RX) 316, the received high-frequency signal is mixed with the receiving local oscillation signal uttered from the frequency synthesizer (SYN) 313, converted into an intermediate frequency signal, and then demodulated into a received voice signal. The received voice signal is amplified by the low-frequency amplifier 317 and then output from a speaker 318 as a receiver.

FIG. 4 is a block diagram showing a digital transmission-compatible radio telephone according to a second embodiment of the present invention.

In FIG. 4, in the radio telephone compatible with digital transmission according to the second embodiment, the same components as those in the second embodiment shown in FIG. In the second embodiment, the output from the voice extraction circuit 12 is encoded into a predetermined code, and the transmission circuit (T
X) an encoding circuit 401 for sending to 13 and a receiving circuit 31
And a decoding circuit 402 for demodulating the coded signal received from 6 into an analog signal and sending it to the low frequency intensifier 317.

Next, with regard to the operation of the second embodiment and the method of extracting and transmitting the speaker's voice, the transmitting side replaces the D / A 323 in the first embodiment with encoding in the second embodiment. The circuit 401 is connected, and on the receiving side, RX3
The decoding circuit 402 is connected between the low-frequency amplifier 16 and the low-frequency amplifier 317, and the other configuration does not change.

As described above, in the present embodiment, the speaker inputs a voice to the voice extraction circuit 12 and registers the voiceprint of the speaker in advance. Next, the speaker's voice is extracted from the signal including the speaker's voice collected from the microphone 11 and the speaker's voiceprint registered in advance, and ambient noise is removed.

This eliminates the need for a microphone for capturing ambient noise, and thus eliminates the need for a plurality of microphones in the present embodiment.

[0040]

As described above, according to the present invention, prior to a call, a speaker's voiceprint is registered in advance, and a comparison is made with a registered speaker's voiceprint from a signal collected from a transmitter during a call. Providing the speaker voice extracting / transmitting means for extracting and transmitting the voice of the speaker has the following effects.

The first effect is that the microphone for collecting ambient noise and its circuit are not required, and the size and weight can be reduced.

The reason is that the voiceprint of the speaker is registered in advance,
Since the ambient noise is removed from the transmission signal collected by one transmitter based on the information, a microphone for collecting ambient noise and their circuits are not required.

The second effect is that the S / N can be improved as compared with the related art.

The reason is that, in the prior art, the distance between the microphone that collects ambient noise and the microphone that collects the voice of the speaker affected the S / N ratio. , The voice signal of the speaker is extracted from the voiceprint of the speaker, so that the S / N can be improved compared to the related art.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a voice extraction circuit according to the present embodiment.

FIG. 3 is a block diagram illustrating a wireless telephone supporting analog transmission according to a first example of the present embodiment.

FIG. 4 is a block diagram illustrating a digital transmission-compatible wireless telephone according to a second example of the present embodiment.

FIG. 5 is a block diagram showing an example of a conventional telephone.

[Explanation of symbols]

 Reference Signs List 11 microphone 12 sound extraction circuit 13 transmission circuit (TX) 14 antenna 22 switch 23 comparison filter 24 voiceprint analysis circuit 26 voiceprint memory 311 console unit (CU) 312 control circuit (CONT) 313 frequency synthesizer (SYN) 314 duplexer (DUP) 316 Receiving circuit (RX) 317, 321 Low frequency amplifier 318 Speaker 322 A / D converter (A / D) 323 D / A converter (D / A) 401 Encoding circuit 402 Decoding circuit

Claims (6)

[Claims] 1. Prior to a call, after registering a voiceprint of a speaker in advance, a voice of the speaker is extracted by comparing the voiceprint of the registered speaker with a signal collected from a transmitter during a call. A communication function device comprising a speaker voice extraction / transmission means for transmitting and transmitting. 2. Prior to a call, a voiceprint of the call is registered in advance, and at the time of a call, the voiceprint of the registered speaker is extracted from a signal collected from a transmitter and the voiceprint of the registered speaker is extracted. A speaker function extracting / transmitting method for a call function device, characterized in that: 3. A telephone function device comprising a voiceprint registration means for registering a voiceprint of a speaker before a call. 4. A speaker voice passing filter means for making a voice of the speaker talk by comparing a voiceprint of the speaker registered in advance with a signal collected from a transmitter during the talk of the speaker. A call function device characterized by the above-mentioned. 5. A switching unit for switching between voiceprint registration and communication, and a voiceprint analysis unit for analyzing the speaker's voiceprint from a speaker's voice from a transmitter when the switching unit is switched to the voiceprint registration side. A voiceprint storage unit that stores the voiceprint of the speaker analyzed by the voiceprint analysis unit; and, when the switching unit is switched to the call side, a signal collected from the transmitter is transmitted from the voiceprint storage unit. A communication function device comprising: a voice-speech-passing filter unit that extracts the voice of the speaker by comparing with the voiceprint of the speaker. 6. A switching section for switching between voiceprint registration and communication is set on the voiceprint registration side prior to a call, and the voiceprint of the speaker is analyzed by a voiceprint analysis section from a speaker's voice from a transmitter. From the result of this analysis, the voiceprint of the speaker is stored in the voiceprint storage unit, and then the switching unit is set to the call side, and the signal collected from the transmitter and the voiceprint storage unit are output from the voiceprint storage unit. Comparison with the speaker's voiceprint is performed by the speaker voice call filter unit,
As a result, a method of extracting and transmitting a speaker's voice in a telephone function device, wherein the speaker's voice is extracted and transmitted.