JP2000261529A - Speech unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an optimum transmission signal by changing the mixing ratio of two frequency band signals comprising a high frequency component and a low frequency component of a signal picked up by an air conduction microphone to a frequency band signal including the low frequency component of a signal picked up by a bone conduction microphone and putting together the three frequency band signals on the basis of an estimated surrounding noise level. SOLUTION: Filters 6, 7 receive a signal picked up by an air conduction microphone 1 to separate the signal into two frequency band signals consisting of a high frequency component whose frequency is 1 kHz or over and a low frequency component whose frequency is lower than 1 kHz. A filter 8 receiving a signal picked up by a bone conduction microphone 2 extracts a frequency band signal including a low frequency component whose frequency is 1 kHz or below. A transmission state detection section 9 and a control information provision section 10 estimate a surrounding noise level from the signal picked up by the air conduction microphone to control level control sections 11, 12. Control section 11, 12 change the mixing ratio of the high frequency band signal from the filter 6 to each of the low frequency band signals from filters 7, 8 and a mixing circuit 13 mixes the signals at this mixing rate. Thus, a proper transmission signal to a noise level can be obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus which uses an air-conducting microphone and a bone-conducting sound pickup (hereinafter referred to as "bone-conducting microphone") as a transmitter, and which is attached to an ear for talking. In low noise to high noise environments,
The present invention relates to a communication device that transmits a good voice signal.

[0002]

2. Description of the Related Art There has been conventionally provided a communication device capable of transmitting and receiving a telephone by integrally attaching a transmitter and a receiver to an ear, and using a gas-conducting microphone as a transmitter of the communication device. And bone conduction microphones. The sound pickup signal of the air conduction microphone has good sound quality in a wide band, but is easily affected by ambient noise. On the other hand, the sound pickup signal of the bone conduction microphone has a narrow band (only low frequency components) and poor sound quality. And is less susceptible to ambient noise. Conventionally, a communication apparatus that processes a signal by dividing the signal into a low-frequency component and a high-frequency component using these features has been proposed. That is,
The user manually switches the switch according to the presence or absence of the surrounding noise level, and according to the switch, the high frequency component of the sound pickup signal of the air conduction microphone and the low frequency of the signal collected by the bone conduction microphone are changed. It does or does not perform mixing with the band frequency component. However, since the user manually switches the switch, the two signals are not mixed or not, so that there is a disadvantage that the switch switching operation is complicated for the user.

To compensate for this drawback, ambient noise is detected, and the mixing ratio between the high-frequency component of the sound pickup signal of the air-conducting microphone and the low-frequency component of the signal picked up by the bone-conducting microphone is optimized. As described above, a method for automatically determining the mixing ratio has been proposed (Japanese Patent Application No. 6-103766).

[0004]

However, in both of the above prior arts, when there is no or little ambient noise, the emphasis is placed on securing the frequency band, and the bone conduction sound is left. There is a drawback that the bone-conducted sound appears as a component signal in the bone-conducted sound, and the poor quality of the bone-conducted sound is conspicuous. On the other hand, if the bone conduction sound is removed with emphasis on sound quality, there is a disadvantage that a low frequency band cannot be secured.

[0005] In order to overcome this drawback, a low-frequency component and a high-frequency component in the sound pickup signal of the air-conducting microphone are generated in order to generate a good transmission signal regardless of the presence or absence of the ambient noise. A method of synthesizing a transmission signal using a low-frequency component of a sound pickup signal of a bone conduction microphone has been proposed (Japanese Patent Application No. 8-168895).

However, as in this conventional example, the low frequency component and the high frequency component of the sound signal picked up by the air-conducting microphone and the low frequency component of the sound signal picked up by the bone-conducting microphone are simply synthesized. In this method, it is not possible to synthesize an optimal transmitted signal regardless of the presence or absence of transmitted voice, and it is not possible to synthesize an optimal transmitted signal according to each noise level. The present invention provides a communication device capable of synthesizing an optimum transmission signal irrespective of the presence or absence of a transmission voice, and synthesizing an optimum transmission signal in accordance with each noise level. The purpose is to do so.

[0007]

SUMMARY OF THE INVENTION The present invention provides an air-conducting microphone, a bone-conducting microphone, and a band signal including a low-frequency component of about 1 KHz or less in a sound pickup signal of the air-conducting microphone.
Signal extraction for extracting a band signal including a high frequency component of approximately 1 KHz or more in the sound signal collected by the air conduction microphone and a band signal including a low frequency component of approximately 1 KHz or less in the sound signal collected by the bone conduction microphone. Means, an ambient noise level estimating means for estimating the ambient noise level, and an ambient noise level estimated by the ambient noise level estimating means,
A band signal including a low-frequency component in the sound signal collected by the air-conducting microphone; a band signal including a high-frequency component in the sound signal collected by the air-conducting microphone; and a low-frequency signal in the sound signal collected by the bone-conducting microphone. A communication apparatus comprising: a band signal combining unit that changes a mixing ratio with a band signal including a component, combines the three band signals, and outputs the combined signal as a transmission signal.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram showing a communication apparatus 100 according to a first embodiment of the present invention.

The communication device 100 includes an air conduction microphone 1, a bone conduction microphone 2, an air conduction microphone amplifier 3, a bone conduction microphone amplifier 4, a filter 5, and a high frequency component of the air conduction microphone 1. Filter 6, filter 7 for low-frequency components of air conduction microphone 1, filter 8 for low-frequency components of bone conduction microphone 2, transmission state detection unit 9, control information adding unit 10, level control unit 11 and 12, a mixing circuit 13, and an output terminal 14 to a telephone line circuit.

The filter 5 is a filter that divides a sound signal output from the air conduction microphone amplifier 3 into a band signal containing a low frequency component of 1 KHz or less and a band signal containing a Takagi frequency component of 1 KHz or more.

The high-frequency component filter 6 of the air-conducting microphone 1 has a frequency of 1 KH in the collected signal output by the air-conducting microphone 1.
This is a filter that passes a band signal including a high-frequency component of z or more. The low-frequency component filter 7 of the air-conducting microphone 1 is used to detect 1K of the picked-up signal output by the air-conducting microphone 1.
This is a filter that passes a band signal including a low-frequency component below Hz. The filter 8 for the low-frequency component of the bone conduction microphone 2 is used to detect the 1
This is a filter that passes a band signal containing a low frequency component of KHz or less.

The transmission state detector 9 detects the level of the sound pickup signal of the air conduction microphone 1 output by the air conduction microphone amplifier 3 and the sound pickup signal of the bone conduction microphone 2 output by the bone conduction microphone amplifier 4. The level is compared with the level to determine the presence or absence of the transmitted voice.

The control information providing unit 10 includes a transmission state detecting unit 9
The level control unit 11 of the band signal including the high frequency component of 1 KHz or more based on the voice presence / absence determination information
It provides control information on the presence or absence of voice to the level control unit 12 for a band signal including a low frequency component of 1 KHz or less.

The level control unit 11 determines the level of a band signal containing a high-frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1 based on the control information about the presence or absence of sound output from the control information adding unit 10. It controls the level. The level control unit 12 includes a band signal including a low-frequency component of 1 KHz or less in a sound pickup signal of the air-conducting microphone 1 output from the low-frequency component filter 7 of the air-conducting microphone 1, and a low-frequency signal of the bone-conducting microphone 2. The level of the sound signal collected by the bone conduction microphone 2 output from the frequency component filter 8 and the level of a band signal including a low-frequency component of 1 KHz or less is determined based on the control information about the presence or absence of sound output from the control information providing device 10. , To control the level. Each of the level control units 11 and 12 is configured by, for example, a gain control circuit using a DC voltage. The mixing circuit 13 includes the level control unit 1
1 KHz in the sound pickup signal of the air-conducting microphone 1 output by 1
The band signal including the high frequency component, the sound signal of the air conduction microphone 1 output by the level control unit 12, and the band signal including the low frequency component of 1 KHz or less in the sound signal of the bone conduction microphone 2 are output. This is a circuit that mixes the combined signal with the level. Next, the operation of the communication device 100 will be described. First, the audio signal collected by the air conduction microphone 1 is amplified by the air conduction microphone amplifier 3, and the audio signal collected by the bone conduction microphone 2 is amplified by the bone conduction microphone amplifier 4. . From the sound pickup signal from the air conduction microphone amplifier 3, a band signal including a high frequency component of 1 KHz or more is extracted by the high frequency component filter 6, and a low frequency component of 1 KHz or less is extracted by the low frequency component filter 7. A band signal including a band frequency component is extracted. From the sound pickup signal of the bone conduction microphone 2, a band signal including a low band frequency component of 1 KHz or less is extracted by the low band frequency component filter 8. The transmission state detection unit 9 that determines the presence or absence of the transmission voice compares the level of the sound collection signal of the air conduction microphone 1 with the level of the sound collection signal of the bone conduction microphone 2, and the results are shown in FIGS. 11 and 12. Using the characteristics, the presence or absence of the transmitted voice is determined. That is, FIG.
When there is no transmission sound, the relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level is shown. When there is no transmission signal, only the noise component is transmitted to the air conduction microphone 1. As the noise hardly enters the bone conduction microphone 2, the level of the sound pickup signal of the air conduction microphone 1 increases according to the noise level, and the level of the sound pickup signal of the bone conduction microphone 2 depends on the noise level. Do not stay low. On the other hand, FIG. 12 shows a relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level when a transmission sound is present. Since the transmission signal enters the air-conducting microphone 1 and the bone-conducting microphone 2, the gain is set so that the level of the signal collected by the bone-conducting microphone 2 is higher than the level of the signal collected by the air-conducting microphone 1. By doing so, as shown in FIG.
The level of the signal collected by the bone-conducting microphone 2 can always be increased. Therefore, by comparing the level of the signal collected by the bone-conducting microphone 2 with the level of the signal collected by the air-conducting microphone 1, the transmission is performed. The presence / absence of speech voice can be identified.

When the transmission state detection unit 9 for determining the presence or absence of the transmission voice receives the determination information that the transmission is not performed, the control information adding unit 10 converts the sound pickup signal of the air conduction microphone 1 into noise. Consider and integrate the signal to estimate the noise level. On the other hand, when receiving the determination information that the user is talking, the noise level estimated immediately before is held.

The signal level control unit 11 adjusts the level of a band signal containing a high frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1. From the noise level estimated based on the control information from the control information adding unit 10, FIG.
The level of a band signal including a high-frequency component of 1 KHz or more in the air conduction microphone 1 is controlled based on a relationship with a preset noise level using 0.

That is, FIG. 10 shows the relationship between the optimum level and the ambient noise level in a band signal containing a high-frequency component of 1 KHz or more in the collected air conduction sound. As shown in FIG. 10 of the microphone 1, the level of a band signal including a high-frequency component of 1 KHz or more in the picked-up signal is set to 1KH in the audio signal based on a predetermined weight so that the sound quality becomes optimum.
The level of a band signal containing a high frequency component equal to or higher than z is adjusted. Signal level control is performed on the level of a band signal containing a low-frequency component of 1 KHz or less in the sound pickup signal of the air-conducting microphone 1 and a band signal containing a low-frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2. The part 12 adjusts,
Then they are combined. From the noise level estimated based on the control information from the control information providing unit 10, using FIG. 8, based on a preset relationship between the noise level and the weight,
The level of a band signal including a low-frequency component of 1 KHz or less in the pickup signals of the microphones 1 and 2 is controlled.

A band signal containing a high-frequency component of 1 KHz or more in the transmission signal from the signal level control unit 11;
1 KHz in the transmission signal from the signal level control unit 12
The following band signals including the low frequency components are mixed by the mixing circuit unit 13.

FIG. 2 is a block diagram showing a specific example of the control information adding unit 10 in the communication device 100. The control information providing unit 10 includes a rectifying unit 10-1 and a switch 10-2.
And a noise level holding unit 10-3. Rectifier 10
-1 indicates that the picked-up signal output from the amplifier 3 of the air-conducting microphone 1 is an AC signal, which is subsequently processed by a DC signal.
The switch 10-2 rectifies the output signal of the amplifier 3 of the air conduction microphone 1 with a diode or the like.
A switch for turning on / off the connection of the rectified signal to the noise level holding unit 10-3 based on the device that determines the presence or absence of the transmission voice. The noise level holding unit 10-3 holds the output signal of the switch 10-2 at a predetermined time constant, and controls the level control unit 11 of a band signal including a high frequency component of 1 KHz or more and a low level of 1 KHz or less. This is a holding unit that provides, to the level control unit 12 of the signal of the band signal including the band frequency component, as control information on the presence or absence of voice. In addition,
The time constant of the noise level holding unit 10-3 is, for example, 1 second or less.

Next, the operation of the control information providing unit 10 will be described.

When the rectifying section 10-1 rectifies the output signal of the air-conducting microphone 1 and the transmitting state detecting section 9 determines that the transmitting state is not transmitting, the rectifying section 10-1 converts the rectified signal into a switch 10 signal. -2 is sent to the noise level holding unit 10-3. Conversely, when the transmission state detection unit 9 determines that the telephone is transmitting, the switch 10-2 interrupts the connection to the noise level holding unit 10-3.

The noise level holding unit 10-3 holds the output of the switch 10-2 at a predetermined time constant, and
The control information relating to the presence / absence of voice is provided to a level control unit 11 for a band signal including a high frequency component of not less than 1 Hz and a level control unit 12 for a signal of a band signal including a low frequency component not higher than 1 KHz.

FIG. 3 is a block diagram showing a communication apparatus 200 according to a second embodiment of the present invention.

The communication device 200 is basically the same as the communication device 100, except that the transmission state detecting filter 15,
The communication device 100 is different from the communication device 100 in that the communication device 100 includes a control information providing unit 10 a instead of the control information providing unit 10.

The transmitting state detecting filter 15 is a filter for detecting the transmitting state of the air-conducting microphone 1, and is a filter for passing a band signal output from the air-conducting microphone 1 and containing a low-frequency component of 1 KHz or less. is there. The transmission state detection filter 16 is a filter that detects the transmission state of the bone conduction microphone 2 and is a filter that passes a band signal including a low-frequency component of 1 KHz or less output by the bone conduction microphone 2.

The transmission state detection unit 9 for determining the presence or absence of the transmission voice is the 1 KHz signal output from the transmission state detection filter 15.
FIG. 1 compares the level of the band signal containing the following low-frequency components with the level of the band signal containing the low-frequency components of 1 KHz or less output by the transmission state detection filter 16.
1. The presence / absence of transmitted voice is determined using the characteristics shown in FIG. That is, as shown in FIG. 11, the level of the sound pickup signal of the air-conducting microphone 1 increases in accordance with the noise level, and the level of the sound pickup signal of the bone-conducting microphone 2 does not depend on the noise level.
Stays low. On the other hand, by setting the gain so that the level of the sound pickup signal of the bone conduction microphone 2 is higher than the level of the sound pickup signal of the air conduction microphone 1, as shown in FIG. 2 can be increased. Therefore, by comparing the level of the collected signal of the bone-conducting microphone 2 with the level of the collected signal of the air-conducting microphone 1, it is possible to determine the presence or absence of the transmitted voice. Can be identified.

FIG. 4 is a block diagram showing the control information adding unit 10a in the communication device 200.

The control information adding unit 10a includes noise estimation filters 10-4H and 10-4L, and rectifiers 10-1H and 10-1H.
0-1L, switches 10-2H and 10-2L, and noise level holding units 10-3H and 10-3L.

Here, it is assumed that the time constant of the noise level holding unit 10-3H is, for example, one digit smaller than the time constant of the noise level holding unit 10-3L.
Assume that the time constant of -3H is, for example, 0.1 second or less, and the time constant of the noise level holding unit 10-3L is, for example, 1 second or less. The noise estimation filter 10-4H is a filter that passes a band signal including a high frequency component of the signal output from the amplifier 3 of the air conduction microphone 1, and is a filter for estimating the noise of the air conduction microphone 1. This is a filter for a band frequency component. The noise estimation filter 10-4L is a filter that passes a band signal including a low-frequency component of the signal output from the amplifier 3 of the air-conducting microphone 1, and is a low-pass filter for estimating the noise of the air-conducting microphone 1. This is a filter for a band frequency component. H and L in the reference numerals 10-1, 10-2, and 10-3 are codes indicating a high frequency component and a low frequency component, respectively.

The control information providing unit 10a estimates the noise level of the high-frequency component band, and sends the estimated noise level to the level control unit 11 as control information. Also,
The control information adding unit 10a estimates the noise level of the low frequency component band, and sends the estimated noise level to the level control unit 12 as control information.

In communication device 200, if the required filter characteristics are equal, filters can be shared, and the number of filters can be reduced. For example, when the frequency characteristics of the low-frequency component filter 8 and the transmission state detection filter 16 are equal, the output signal of the low-frequency component filter 8 is branched to the transmission state detection unit 9 to transmit the signal. The talking state detecting filter 16 can be omitted.

FIG. 5 shows the mixing circuit 1 in the above embodiment.
FIG. 3 is a block diagram showing a combined signal processing unit 20 provided between a transmission signal output terminal 3 and a transmission signal output terminal 14.

When the frequency band of the transmission signal is limited to the frequency band of the communication device, the synthetic signal processing unit 20 corrects the frequency characteristics and corrects the level. It has a level correcting device 18 and a device 19 for limiting the signal band to the frequency band of the communication device.

The frequency characteristic correcting device 17 includes a band signal including a low-frequency component of 1 KHz or less in the sound pickup signal of the air-conducting microphone 1 and a 1K signal in the sound pickup signal of the air-conducting microphone 1.
This is to correct the frequency characteristics of the signal after combining a band signal including a high-frequency component of not less than 1 Hz and a band signal including a low-frequency component of not more than 1 KHz in the sound pickup signal of the bone conduction microphone 2. The level correcting device 18 corrects the level of the synthesized signal. The device 19 for limiting the signal band to the frequency band of the communication device limits the frequency band of the synthesized signal to the frequency band of the communication device, and can be selected as necessary. Note that the arrangement of the frequency characteristic correction device 17, the level correction device 18, and the device 19 for limiting the signal band to the frequency band of the communication device may be interchanged.

FIG. 6 shows the mixing circuit 1 in the above embodiment.
FIG. 3 is a block diagram showing a combined signal processing unit 21 provided between the transmission signal output terminal 3 and a transmission signal output terminal 14;

The synthetic signal processing section 21 corrects the frequency characteristics and corrects the level when the frequency band of the transmission signal is limited to the frequency band of the communication device. Level correction device 18.

When the synthesized signal processing unit 21 is used in place of the synthesized signal processing unit 20, the function of the device 19 for limiting the frequency band of the synthesized signal to the frequency band of the communication device is changed to the filters 6, 7, and 8. It is necessary to have. That is, the filter 6 when the synthesized signal processing unit 21 is used,
Reference numerals 7 and 8 denote band-pass filters having the frequency band of the communication device as a pass band.

FIG. 7 shows the relationship between the sound quality of a band signal containing a low-frequency component of 1 KHz or less and the ambient noise level in the sound signal picked up by the air-conducting microphone 1 and the sound signal picked up by the bone-conducting microphone 2 below 1 KHz. FIG. 4 is a diagram illustrating a relationship between sound quality of a band signal including a band frequency component and an ambient noise level.

1 KHz in the sound pickup signal of the air conduction microphone 1
The sound quality of the band signal including the following low frequency components is greatly affected by the level of the ambient noise. When the noise level is high, the sound quality is significantly deteriorated. On the other hand, the sound quality of the band signal including the low frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2 is relatively unaffected by the level of the ambient noise, and even when the noise level is large, the sound quality deteriorates. Relatively small.

FIG. 8 is a diagram showing the relationship between the optimum mixing ratio and the ambient noise level in a band signal containing a low-frequency component of 1 KHz or less in the collected bone conduction sound and air conduction sound.

Therefore, a band signal including a low-frequency component of 1 KHz or less in the sound pickup signal of the air-conducting microphone 1 is:
By controlling the weight at the time of addition with the band signal including the low frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2 in accordance with the noise level as shown in FIG. To synthesize a band signal including a low frequency component of 1 KHz or less.

On the other hand, since the frequency component of the sound signal picked up by the bone-conducting microphone is only the low-frequency component, the band signal containing the high-frequency component of 1 KHz or more in the picked-up signal is:
Only the high frequency component of the sound pickup signal of the air conduction microphone 1 is included.

In FIG. 8, when the mixture ratio of the air conduction sound and the mixture ratio of the bone conduction sound become equal, the noise level is, for example, 60 to 80 dB.

FIG. 9 is a diagram showing the relationship between the sound quality of a band signal containing a high-frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1 and the ambient noise level.

1 KHz in the sound pickup signal of air conduction microphone 1
The sound quality of the band signal including the high frequency components described above is greatly affected by the level of the ambient noise, and when the noise level is high, the sound quality is significantly deteriorated. However, since the band signal including the high frequency component of 1 KHz or more in the collected signal is only the high frequency component of the collected signal of the air-conducting microphone 1, from the viewpoint of securing the band, the high frequency component is The levels of the components should be as high as possible. That is, level correction is required.

FIG. 10 shows 1KH in the collected air conduction sound.
FIG. 9 is a diagram illustrating a relationship between an optimum level and an ambient noise level in a band signal including a high-frequency component equal to or more than z.

Therefore, depending on the noise level, the level of the band signal including the high-frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1 is adjusted based on a predetermined weight so that the sound quality becomes optimum. 1KH in signal
The level of a band signal containing a high frequency component equal to or higher than z is adjusted as shown in FIG.

In FIG. 10, the noise level at the connection between the horizontal line of the characteristic and the straight line descending to the right is, for example, 60 to 80 dB.

In determining the weights described above, it is necessary to obtain an estimated value of the level of the ambient noise.

FIG. 11 is a diagram showing the relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level when there is no transmission sound.

The relationship between the magnitude of the signal collected by the bone-conducting microphone 2 and the magnitude of the signal collected by the air-conducting microphone 1 can be set as shown in FIGS. That is, when there is no transmission signal, only a noise component enters the air-conducting microphone 1 and almost no noise enters the bone-conducting microphone 2. Therefore, the level of the sound pickup signal of the air-conducting microphone 1 increases according to the noise level, and the level of the sound pickup signal of the bone-conducting microphone 2 does not depend on the noise level and remains low.

FIG. 12 is a diagram showing the relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level when a transmission sound is present.

When a transmission signal exists, the transmission signal enters the air conduction microphone 1 and the bone conduction microphone 2. At this time, by setting the gain so that the level of the sound pickup signal of the bone-conducting microphone 2 is higher than the level of the sound pickup signal of the air-conducting microphone 1, the bone is always set as shown in FIG. The level of the sound pickup signal of the microphone 2 can be increased. Therefore, by comparing the level of the sound pickup signal of the bone-conducting microphone 2 with the level of the sound pickup signal of the air-conducting microphone 1, it is possible to identify the presence or absence of the transmitted voice.

The level of the bone-conducting microphone 2 is compared with the level of the air-conducting microphone 1, and the level of the picked-up signal of the bone-conducting microphone 2 is
When the level is smaller than the level of the sound pickup signal of the air-conducting microphone 1,
It can be determined that there is no transmission voice. In this case, air conduction microphone 1
Is regarded as ambient noise, and by integrating this signal, the ambient noise level can be estimated.

Conversely, when the level of the sound pickup signal of the bone-conducting microphone 2 is higher than the level of the sound pickup signal of the air-conducting microphone 1, it is in a transmission state. In this case, since the sound is superimposed on the sound pickup signal of the air-conducting microphone 1 and input, the level immediately before the transmission state is set as the noise level so that this level is not erroneously determined as the noise level.

Further, the level of a band signal containing a low frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2 and the level of a band signal containing a low frequency component of 1 KHz or less in the sound pickup signal of the air conduction microphone 1 are shown. By comparing the level with the level, it is also possible to consider the difference between the frequency band of the sound collection signal of the bone conduction microphone 2 and the frequency band of the sound collection signal of the air conduction microphone 1.

Further, 1 in the sound pickup signal of the air-conducting microphone 1
A band signal containing a low-frequency component of 1 KHz or less, a band signal containing a high-frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1, and a low-frequency signal of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2. 1 KHz when changing the level with the band signal including the component according to the noise level
The following mixing ratio of the band signal including the low frequency component is based on the ambient noise level estimated from the band signal including the low frequency component of 1 KHz or less in the sound pickup signal of the air-conducting microphone 1, and the high frequency range of 1 KHz or more is obtained. If the mixing ratio of the band signal including the frequency component is determined based on the ambient noise level estimated from the band signal including the high frequency band number component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1, the mixing ratio of the band signal can be further increased. Control accuracy is improved.

1 KHz in the sound pickup signal of air conduction microphone 1
A band signal containing the following low-frequency components, a band signal containing a high-frequency component of 1 KHz or more, and a band signal containing a low-frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2 are converted into noise levels. Since the frequency characteristics and the level of the composite signal change when the mixing ratio is changed and added according to the above, the frequency characteristics and the level are corrected. Also, by limiting the frequency band to the frequency band of the communication device, it is possible to prevent unnecessary noise from being generated in the transmission signal.

As described above, the sound quality of the band signal including the low frequency component of 1 KHz or less in the sound pickup signal of the air conduction microphone 1 shown in FIG.
The sound quality of the band signal containing low frequency components below KHz,
It is possible to effectively use the relationship with the ambient noise level, the sound quality of the band signal including the high frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1 shown in FIG. it can.

In accordance with the noise level, a band signal containing a low-frequency component of 1 KHz or less in the sound pickup signal of the air conduction microphone 1 and a band containing a low-frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2. By controlling the mixing ratio with the signal based on the weights in FIG. 8, a band signal including a low-frequency component of 1 KHz or less can be automatically synthesized.

Further, by controlling the level of the band signal including the high frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1 based on the weight in FIG. 10, the high frequency component of 1 KHz or more is controlled. Band signal containing
Can be synthesized automatically.

Therefore, the voice signal can be automatically synthesized in accordance with the noise level over the entire transmission voice band. Further, by correcting the frequency characteristics and the level of the synthesized signal and limiting the frequency band to the frequency band of the communication device, the quality of the transmission signal can be improved.

In the above embodiment, "approximately 1 KHz" may be assumed instead of "1 KHz". That is, instead of 1 KHz, for example, 800 Hz to 1
200 Hz may be assumed. That is, in each of the above embodiments, the air-conducting microphone 1 and the bone-conducting microphone 2
A band signal containing a low-frequency component of approximately 1 KHz or less, a band signal containing a high-frequency component of approximately 1 KHz or more, and a signal of the air-conducting microphone 1. Approximately 1K
And a band signal containing a low-frequency component of less than 3 Hz so as to obtain a transmission signal having an optimum sound quality according to the ambient noise level. By comparing and judging the collected sound signal, the presence / absence of the transmitted voice is automatically determined so that the transmitted voice level is not erroneously determined as the ambient noise level.

According to the above embodiment, in an apparatus for communicating using an air-conducting microphone 1 and a bone-conducting microphone 2 as a transmitter,
It enables a good voice call in the usage environment. By comparing the level of the sound pickup signal of the air-conducting microphone 1 with the level of the sound pickup signal of the bone-conducting microphone 2, the presence or absence of the transmitted voice is automatically determined. When it is determined that there is no transmission voice, the ambient noise level is estimated using the sound pickup signal of the air conduction microphone 1. If it is determined that there is a transmitted voice, the voice is superimposed on the picked-up signal of the air-conducting microphone 1, and the level immediately before the transmitting state is changed so that the level is not erroneously determined as the noise level. Noise level. A band signal including a low-frequency component of approximately 1 KHz or less in a sound pickup signal of the air-conducting microphone 1 according to the estimated ambient noise level;
A band signal including a high frequency component of about 1 KHz or more;
A transmission signal is generated by synthesizing a low-frequency component of approximately 1 KHz or less in a sound pickup signal of the bone conduction microphone 2 so as to have an optimum sound quality. As described above, it is possible to accurately reflect the level of the ambient noise, synthesize an optimal transmission signal according to each noise level regardless of the presence or absence of the transmission voice, and improve the sound quality of the transmission signal.

Although the above-described embodiment is a communication device, it is possible to generate a transmission signal so that optimum sound quality can be obtained even in a general telephone call in the environment and conditions of ambient noise during use. become.

[0066]

According to the present invention, an optimum transmission signal can be synthesized irrespective of the presence or absence of a transmission voice, and an optimum transmission signal can be synthesized according to each noise level. This has the effect that it can be performed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a communication device 100 according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a specific example of a control information adding unit 10 in the communication device 100.

FIG. 3 is a block diagram showing a communication device 200 according to a second embodiment of the present invention.

FIG. 4 is a control information providing unit 10a in the communication device 200.
FIG.

FIG. 5 is a block diagram showing a composite signal processing unit 20 provided between the mixing circuit 13 and the transmission signal output terminal 14 in the embodiment.

FIG. 6 is a block diagram showing a combined signal processing unit 21 provided between the mixing circuit 13 and the transmission signal output terminal 14 in the embodiment.

FIG. 7 shows the relationship between the sound quality of a band signal containing a low-frequency component of 1 KHz or less and the ambient noise level in the sound pickup signal of the air-conducting microphone 1 and 1KH in the sound pickup signal of the bone-conducting microphone 2;
FIG. 7 is a diagram illustrating a relationship between a sound quality of a band signal including a band frequency component equal to or less than z and an ambient noise level.

FIG. 8 is a diagram showing a relationship between an optimum mixing ratio and an ambient noise level in a band signal including a low-frequency component of 1 KHz or less in the collected bone conduction sound and air conduction sound.

FIG. 9 is a diagram showing a relationship between a sound quality of a band signal including a high frequency component of 1 KHz or more in a sound pickup signal of the air-conducting microphone 1 and an ambient noise level.

FIG. 10 is a diagram illustrating a relationship between an optimal level and an ambient noise level in a band signal including a high-frequency component of 1 KHz or more in a collected air-conducted sound.

FIG. 11 is a diagram showing a relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level when there is no transmission sound.

FIG. 12 is a diagram showing the relationship between the level of the collected bone conduction sound, the level of the air conduction sound, and the ambient noise level when a transmission sound is present.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Air conduction microphone, 2 ... Bone conduction microphone, 3 ... Air conduction microphone amplifier, 4 ... Bone conduction microphone amplifier, 5 ... Low frequency component below 1 KHz A filter for dividing a band signal containing a high-frequency component of 1 KHz or more into a band signal containing a high-frequency component of 1 KHz or more; 6 a filter for passing a band signal containing a high-frequency component of 1 kHz or more for an air-conduction microphone; 8. A filter that passes a band signal containing the following low frequency components: 8: a filter for a bone conduction microphone that passes a band signal containing a low frequency component of 1 KHz or less, 9: a transmission state detection unit, 10: control information Providing section, 11: a level control section of a band signal including a high-frequency component of 1 KHz or more in the sound pickup signal of the air-conducting microphone 1, 12 ... Level control unit for a band signal containing a low-frequency component of 1 KHz or less and a band signal containing a low-frequency component of 1 KHz or less in the sound pickup signal of the bone conduction microphone 2 from the device 8; 13 ... mixing circuit; Signal output terminal. 10-1, 10-1H, 10-1L: rectifying section, 10-2, 10-2H, 10-2L: electronic switch, 10-3, 10-3H, 103L: noise level holding section, 10-4H: noise High frequency component filter for estimation, 10-4L... Low frequency component filter for noise estimation. Reference numeral 15 denotes a filter for a low-frequency component of a sound signal collected in the air-conducting microphone 1 for detecting a transmission state, and reference numeral 16 denotes a filter for a low-frequency component of a collected signal in the bone conduction microphone 2 for detecting a transmission state. 17: frequency characteristic correction device, 18: level correction device, 19: device for limiting the signal band to the frequency band of the communication device.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tetsuma Sakurai 3-19-2 Nishishinjuku, Shinjuku-ku, Tokyo Japan Telegraph and Telephone Corporation (72) Inventor Yutaka Nishino 3-19, Nishishinjuku, Shinjuku-ku, Tokyo No. 2 Nippon Telegraph and Telephone Corporation F term (reference) 5D017 BA01 5K023 BB09 EE05 EE06 5K027 BB03 DD12 DD16 9A001 HH16 KK13 KK56

Claims (7)

[Claims] An air-conducting microphone; a bone-conducting microphone; a band signal including a low-frequency component of approximately 1 KHz or less in a sound-collecting signal of the air-conducting microphone, and approximately 1 kHz or more in a sound-collecting signal of the air-conducting microphone. And a band signal containing a high frequency component of about 1 KHz in the sound pickup signal of the bone conduction microphone.
Signal extracting means for extracting a band signal including the following low-frequency components; ambient noise level estimating means for estimating the ambient noise level; A band signal including a low-frequency component in the sound pickup signal of the conduction microphone, a band signal including a high-frequency component in the sound pickup signal of the air conduction microphone, and a low-frequency component in the sound collection signal of the bone conduction microphone. A band signal synthesizing means for changing a mixing ratio with the included band signal, synthesizing the three band signals, and outputting as a transmission signal. 2. The apparatus according to claim 1, wherein the ambient noise level estimating means determines the presence or absence of a transmission voice by comparing a sound pickup signal of the air conduction microphone with a sound pickup signal of the bone conduction microphone. Means for determining the presence or absence of transmitted voice; level estimating means for estimating an ambient noise level in accordance with a sound pickup signal of the air-conducting microphone when the determination means determines that transmitted voice is not present; And a surrounding noise level holding unit for holding the ambient noise level estimated by the level estimating unit when the determining unit determines that the communication device is present. 3. The signal extraction unit according to claim 1, wherein the signal extraction unit extracts a band signal including a low-frequency component of approximately 1 KHz or less from a sound pickup signal of the air conduction microphone; Almost 1 KHz from the sound pickup signal of the bone conduction microphone
And a third signal extracting means for extracting a band signal including a low frequency component as described below. 4. The signal extraction unit according to claim 1, wherein the signal extraction unit extracts a band signal including a low-frequency component of approximately 1 KHz or less from a sound pickup signal of the air conduction microphone; Almost 1 KHz from sound pickup signal of air conduction microphone
A second signal extracting means for extracting a band signal containing the high frequency component as described above. 5. The signal extraction means according to claim 1, wherein the signal extraction means extracts a band signal including a low-frequency component of about 1 KHz or less from a sound pickup signal of the air conduction microphone; Almost 1 KHz from sound pickup signal of air conduction microphone
A second signal extracting means for extracting a band signal containing the above high frequency component; and a third signal for extracting a band signal containing a low frequency component of about 1 KHz or less from the sound pickup signal of the bone conduction microphone. A communication device, comprising: extraction means. 6. The communication device according to claim 1, wherein a frequency characteristic correction means for correcting a frequency characteristic of the combined signal; a level correction means for correcting a level of the combined signal; And a band pass filter having a band as a pass band. 7. A signal according to claim 1, wherein a frequency characteristic correcting means for correcting a frequency characteristic of the synthesized signal; a level correcting means for correcting a level of the synthesized signal; Frequency band limiting means for limiting a frequency band of an output signal to a frequency band of the telephone device.