JP5547432B2 - Automatic volume control device - Google Patents

Description

  The present invention relates to an automatic volume control device, and more particularly, by reducing only the monaural component of the audio band according to the utterance when the conversation is performed in a space where music is played, The present invention relates to an automatic volume control device capable of realizing smooth conversation.

  In the interior of a running vehicle, music and radio programs are often played while driving (running). In such a situation, when the driver and the passenger have a conversation, there is a possibility that a smooth conversation (listening of the conversation, etc.) may be hindered by the reproduced sound such as music.

  A general in-vehicle audio apparatus is provided with a volume control switch for adjusting the volume, a mute switch for temporarily reducing the volume, and the like (see, for example, Patent Document 1 and Patent Document 2). . For this reason, the driver or the like often reduces the playback volume of music or the like to such an extent that the conversation is not hindered by operating the volume control switch and the mute switch.

JP 2008-62906 A JP 2006-67490 A

  However, the method of operating the volume control switch every time a conversation is performed to reduce the reproduced sound has a problem that the operation becomes complicated and may hinder smooth conversation. On the other hand, with the method of reducing the music playback sound (output level) using the mute switch, the music playback sound is reduced even when the conversation is interrupted. There was a problem that I could not enjoy it.

  For this reason, it is often the case that a speaker utters a loud voice that can establish a conversation without operating a volume control switch or a mute switch, and the conversation is performed while music is being reproduced. When the conversation continues, there is a problem that the conversation partner may feel tired as well as the speaker.

  The present invention has been made in view of the above problems, and a driver or a passenger does a smooth conversation in a normal state without operating a volume control switch or a mute switch and without generating a loud voice. It is an object of the present invention to provide an automatic sound volume control device that can be used.

  In order to solve the above-described problems, an automatic volume control device according to the present invention includes an audio detection unit that detects an audio signal from an acoustic signal acquired by a microphone, an L channel music signal output from a sound source, and an R signal. Band division means for dividing each of the channel music signals into a first audio band signal composed of a frequency band corresponding to an audio band and a non-audio band signal composed of a frequency band not corresponding to the audio band; Phase inversion means for performing phase inversion for only one of the audio band signals of the first audio band signal for the L channel and the first audio band signal for the R channel, and L that has been phase inverted by the phase inversion means. Either the first audio band signal for the channel or the first audio band signal for the R channel and the phase inverting means A voice band signal synthesizing unit that generates a second voice band signal for the L channel and a second voice band signal for the R channel by synthesizing the other voice band signal that has not been inverted; and the voice detection unit When the audio signal is detected by the above, the second audio band signal for the L channel and the second audio band signal for the R channel generated by the audio band signal synthesis means are used for each channel. When the audio signal is output as an audio band signal and the audio signal is not detected by the audio detection unit, the first audio band signal for the L channel and the R channel for the channel divided by the band division unit Voice band signal selection means for outputting the first voice band signal as it is as a voice band signal for each channel, and output by the voice band signal selection means The L-channel audio band signal and the R-channel audio band signal, and the L-channel non-audio band signal and the R-channel non-audio band signal that are band-divided by the band dividing unit, respectively. It is characterized by comprising music signal synthesizing means for generating a music signal for the entire band for the L channel and a music signal for the entire band for the R channel by combining each channel.

  According to the automatic volume control device of the present invention, each of the L channel music signal and the R channel music signal corresponds to the first audio band signal having a frequency band corresponding to the audio band and the audio band. After the frequency is divided into non-voice band signals composed of non-frequency bands, only one of the voice band signals of the first voice band signal for the L channel and the first voice band signal for the R channel is inverted, Since the other voice band signal that has not undergone phase inversion is synthesized, the synthesized second voice band signal is in a state where the output (signal level) of the monaural component is reduced by synthesis.

  The music signal is synthesized by synthesizing the music signal (second audio band signal) in which the output (signal level) of the monaural component in the audio band is reduced in this way and the non-audio band signal band-divided by the band dividing unit. Because it can effectively reduce only the signal level of the monaural component of the voice band that can hinder conversation while maintaining the overall stereo feeling, enjoy smooth conversation even when music flows It becomes possible.

  In addition, the overall stereo feeling of the music signal (especially the stereo component of the non-speech band signal) is maintained even during a conversation, so the sound quality of the music (realism, power, etc.) can be achieved while realizing a smooth conversation. ) Can be suppressed.

  Furthermore, as described above, in the music signal of the entire band generated by the music signal synthesizing unit, the signal level in the monaural component of the audio band is reduced only when the audio signal is detected by the audio detecting unit. Since it is only when words are spoken by the speaker, the signal level of the monaural component in the audio band of music can be automatically reduced only when conversation is performed.

  In addition, since the volume control of the music signal is performed in response to the detection of the audio signal in this way, in a state where the conversation is not being performed (such as a state where the conversation is interrupted), the monaural component reduction processing (volume control) ) Is not performed. For this reason, in a state where no conversation is performed (a state in which no audio signal is detected), no reduction processing (volume control) is performed on the music signal, so that it is possible to enjoy high-quality music.

  Further, the voice band signal synthesizing unit is configured to provide a channel for each of the second voice band signal for the L channel and the second voice band signal for the R channel synthesized by the voice band signal synthesizing unit. There may be provided gain adding means for weighting the gain in accordance with.

  As described above, in the automatic sound volume control device according to the present invention, the gain addition unit weights the gain corresponding to each channel for the second audio band signal in which the monaural component is reduced. Even if the monaural component is reduced in the audio band signal, the remaining stereo component is emphasized, so that the overall stereo feeling including the audio band (vocal component) can be kept high. Become.

  Further, since the weight addition processing is performed on the second audio band signal by the gain adding means, the second audio band signal for the L channel and the second audio band signal for the R channel become an uncorrelated signal, and the music signal synthesizing means It is possible to prevent the stereo component music signal from interfering when the music signal is synthesized in FIG.

  Further, in the automatic volume control device described above, the band dividing unit extracts each of the music signal for L channel and the music signal for R channel output from the sound source from a frequency band corresponding to a voice band. The first audio band signal is divided into the first audio band signal, the low frequency signal having a frequency band lower than the audio band, and the high frequency signal having a frequency band higher than the audio band, and the first audio band signal And the music signal synthesizing means outputs the audio band signal for the L channel and the audio band signal for the R channel output by the audio band signal selecting means, and the low frequency signal for the L channel. And the low frequency signal for the R channel, the high frequency signal for the L channel, and the high frequency signal for the R channel, respectively. By synthesizing each tunnel, it may be configured to generate a full-band music signals for music signals and R-channel of the total bandwidth of the L-channel.

  Thus, as a non-speech band signal, the audio signal is divided into a low-frequency signal having a frequency band lower than the voice band and a high-frequency signal having a frequency band higher than the voice band. It becomes possible to express the high-band and low-band stereo components of the music signal subjected to the synthesis processing in the synthesis means more effectively.

  Furthermore, by performing phase inversion only for the second audio band signal in the band dividing means in advance, it is possible to reduce the occurrence of interference between the signals when the music signal synthesizing means synthesizes the signals of each band. Is possible.

  According to the automatic volume control device of the present invention, each of the L channel music signal and the R channel music signal is divided into the first voice band signal having a frequency band corresponding to the voice band and the voice band. After band division into non-voice band signals having non-corresponding frequency bands, only one of the voice band signals of the first voice band signal for L channel and the first voice band signal for R channel is inverted in phase. Therefore, the synthesized second audio band signal is in a state in which the output of the monaural component (signal level) is reduced by the synthesis.

  The music signal is synthesized by synthesizing the music signal (second audio band signal) in which the output (signal level) of the monaural component in the audio band is reduced in this way and the non-audio band signal band-divided by the band dividing unit. Because it can effectively reduce only the signal level of the monaural component of the voice band that can hinder conversation while maintaining the overall stereo feeling, enjoy smooth conversation even when music flows It becomes possible.

It is the block diagram which showed schematic structure of the automatic volume control apparatus which concerns on this Embodiment. It is the block diagram which showed schematic structure of the audio | voice emphasis processing part which concerns on this Embodiment. It is the block diagram which showed schematic structure of the audio canceller part which concerns on this Embodiment. It is the block diagram which showed schematic structure of the audio | voice detection part which concerns on this Embodiment. It is the figure which showed the relationship between the audio | voice signal from which the moving average was calculated | required in the moving average part which concerns on this Embodiment, and the audio | voice detection threshold set in an audio | voice detection threshold part. It is the block diagram which showed schematic structure of the 1st zone | band division | segmentation part which concerns on this Embodiment. It is the figure which showed the filter characteristic for producing | generating the low-pass music signal L, the mid-range music signal L, and the high-pass music signal L in the 1st band division part which concerns on this Embodiment. It is the block diagram which showed schematic structure of the volume control part which concerns on this Embodiment. It is the figure which showed as an example the frequency characteristic of the audio | voice signal acquired with the microphone which concerns on this Embodiment. When the white noise for the L channel and the white noise for the R channel are the same monaural signal, the volume control is performed and the volume control is not performed by the automatic volume control device according to the present embodiment. It is the figure which showed the frequency characteristic of white noise. When the white noise for the L channel and the white noise for the R channel are respectively uncorrelated stereo signals, the automatic volume control device according to this embodiment does not perform the volume control. It is the figure which showed the frequency characteristic of the white noise with the case. The figure which showed the frequency characteristic of the music signal with the case where the volume control by an automatic volume control apparatus is not performed when the music signal for L channels and the channel for R has a stereo component. It is.

  Hereinafter, an automatic sound volume control apparatus according to the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a block diagram showing a schematic configuration of the automatic volume control device according to the present embodiment. In the present embodiment, the case where the automatic volume control device 1 is installed in a vehicle will be described as an example. By installing the automatic volume control device 1 according to the present embodiment in a vehicle, it is possible to automatically reduce only the monaural component of the voice band in the music signal output from the sound source according to the presence or absence of conversation. become.

  As shown in FIG. 1, the automatic sound volume control apparatus 1 according to the present embodiment includes a voice enhancement processing unit 2, a voice detection unit (speech detection unit) 3, a first band division unit (band division unit) 4, a second Band division unit (band division unit) 5, volume control unit (phase inversion unit, voice band signal synthesis unit, voice band signal selection unit, gain addition unit) 6, first synthesis unit (music signal synthesis unit) 7, second A synthesizing unit (music signal synthesizing means) 8, a first power amplifier 9, a second power amplifier 10, speakers 11a and 11b, and a microphone M are roughly configured.

[Speech enhancement processor]
First, the speech enhancement processing unit 2 will be described. FIG. 2 is a block diagram showing a schematic configuration of the speech enhancement processing unit 2. The speech enhancement processing unit 2 includes an audio canceller unit 15 and a noise canceller unit 16. The speech enhancement processing unit 2 has a role of reducing noise sounds such as music signals and vehicle running noise from the acoustic signals acquired by the microphone M.

[Audio canceller]
Next, the audio canceller unit 15 will be described. As shown in FIG. 3, the audio canceller unit 15 includes a first bandpass filter unit 21, a second bandpass filter unit 22, a first delay unit 23, a second delay unit 24, and a first adaptive filter unit. 25 and a second adaptive filter unit 26.

  The first band-pass filter unit 21 and the second band-pass filter unit 22 are two-channel music signals output from the sound source, that is, the music signal L for the L channel (left side) and the music signal R for the R channel (right side). On the other hand, by limiting the band of about 200 Hz to 2.6 kHz, it mainly has a role of allowing only the voice band signal among the voice signals to pass.

  The first delay unit 23 and the second delay unit 24 have a role of performing a delay process on the acoustic signal subjected to the band limiting process by the first band pass filter unit 21 and the second band pass filter unit 22. Yes. By the delay processing by the first delay unit 23 and the second delay unit 24, it is possible to correct the propagation delay of the acoustic signal acquired by the microphone M.

  The first adaptive filter unit 25 is roughly configured by a first FIR unit 28, a first LMS unit 29, and a first addition unit 30, and the second adaptive filter unit 26 includes a second FIR unit 31, a second LMS unit 32, a second The adder 33 is roughly configured.

  The first FIR unit 28 includes a finite impulse response filter, and has a function of filtering the acoustic signal collected by the microphone M based on the coefficient control performed by the first LMS unit 29. Yes. The first adder 30 adds the sound signal input from the microphone M in a state where the phase of the music signal L filtered by the first FIR unit 28 is inverted (substantially the microphone). The acoustic signal of the music signal L that has been filtered is subtracted from the acoustic signal of M). The acoustic signal added by the first addition unit 30 is output from the first adaptive filter unit 25 and also output to the first LMS unit 29.

  The first LMS unit 29 uses the acoustic signal (a signal obtained by subtracting the music signal L subjected to the filter processing from the acoustic signal of the microphone M) acquired from the first adding unit 30 and the music signal L from the sound source, Filter coefficient control in the first FIR unit 28 is performed based on the least square algorithm. By providing the first LMS unit 29 in the first adaptive filter unit 25 in this way, the LMS algorithm can be applied.

  Further, as shown in FIG. 3, the second adaptive filter unit 26 includes a first FIR unit 28, a first LMS unit 29, a first adder unit 30, a second FIR unit 31 in the first adaptive filter unit 25, The second LMS unit 32 and the second adder unit 33 are replaced. The configuration and function of the second adaptive filter unit 26 are the same as those of the first adaptive filter unit 25, and thus the description thereof is omitted.

  In the second addition unit 33 of the second adaptive filter unit 26, the music signal R filtered by the second FIR unit 31 with respect to the acoustic signal filtered by the first adaptive filter unit 25 is phase-converted. Is added (substantially, the music signal R filtered by the second FIR unit 31 is subtracted from the acoustic signal filtered by the first adaptive filter unit 25). The acoustic signal added by the second addition unit 33 is output from the second adaptive filter unit 26 to the noise canceller unit 16 and also output to the second LMS unit 32.

  In the audio canceller unit 15, the first adaptive filter unit 25 and the second adaptive filter unit 26 are cascade-connected as shown in FIG. Accordingly, in the audio canceller unit 15, the acoustic signal input from the microphone M in the first adaptive filter unit 25 is subtracted with the music signal L, and then subtracted in the first adaptive filter unit 25 in the second adaptive filter unit 26. The audio signal is subtracted by the music signal R. In this case, since the first adaptive filter unit 25 needs to converge faster than the second adaptive filter unit 26, the adaptive speed is set large. When there are two or more sound sources, the same effect can be obtained by increasing the number of adaptive filter units installed according to the number of channels.

[Noise canceller]
The noise canceller 16 has a role of removing a noise signal from a signal input from the audio canceller 15 using a spectral subtraction method or the like. More specifically, the signal output from the audio canceller 15 includes an audio signal and a noise signal (a signal other than audio). The noise canceller 16 is a temporally stationary component in the frequency spectrum (a component in a situation where there is no speech and a section where there is no speech).
) Is estimated and held as a noise pattern, and at the same time as the noise pattern to be subtracted in the frequency domain of the input signal including the audio signal and the noise signal (signal input from the audio canceller unit 15), the noise pattern By continuing to update itself, the noise signal is removed from the input signal (the signal input from the audio canceller 15).

  As described above, by performing filter processing in the audio canceller unit 15 and further performing noise removal processing in the noise canceller unit 16, music signals, vehicle running noise, and the like can be obtained from the acoustic signals acquired by the microphone M. Since the noise sound is reduced, the voice of the speaker can be emphasized.

[Audio detector]
Next, the voice detection unit 3 will be described. FIG. 4 is a block diagram showing a schematic configuration of the voice detection unit 3. As shown in FIG. 4, the voice detection unit 3 includes an effective value detection unit 41, a moving average unit 42, a voice detection threshold unit 43, and a voice signal hold unit 44.

  The effective value detection unit 41 has a role of detecting an effective value in a predetermined section in the audio signal subjected to the audio enhancement processing by the audio enhancement processing unit 2. The moving average unit 42 has a role of obtaining a moving average of a predetermined section with respect to the signal whose effective value is detected by the effective value detecting unit 41. The voice detection threshold unit 43 has a role of detecting a voice signal whose signal level exceeds the voice detection threshold (threshold) based on a preset voice detection threshold (threshold).

  FIG. 5 is a diagram showing a relationship between the voice signal whose moving average is obtained by the moving average unit 42 and the voice detection threshold set by the voice detection threshold unit 43. When the speech signal is detected by the speech detection threshold unit 43, the speech signal input from the speech enhancement processing unit 2 is in a state where the noise sound is reduced and the speech of the speaker is enhanced by the speech enhancement processing unit 2. Therefore, it is possible to easily detect the audio signal based on the audio detection threshold. In FIG. 5, a part having a signal level higher than the voice detection threshold (for example, a part corresponding to the voice section shown in FIG. 5) is determined as a voice signal detection part.

  The audio signal holding unit 44 has a role of holding a predetermined time for the audio detection signal detected by the audio detection threshold unit 43. The audio signal that is hold-processed by the audio signal holding unit 44 is output to the volume control unit 6 as an audio detection signal.

[First Band Division]
Next, the first band dividing unit 4 will be described. FIG. 6 is a block diagram showing a schematic configuration of the first band dividing unit 4. As shown in FIG. 6, the first band dividing unit 4 includes a first low-pass filter unit 51, a second low-pass filter unit 52, a first high-pass filter unit 53, a second high-pass filter unit 54, and a phase inverting unit 55. And have.

  The first low-pass filter unit 51 and the second low-pass filter unit 52 are configured by two-stage cascade connection of low-pass type third-order Butterworth filters, and the first high-pass filter The unit 53 and the second high-pass filter unit 54 are configured by cascade-connecting high-pass third-order Butterworth filters.

  Using the first low-pass filter unit 51, the second low-pass filter unit 52, the first high-pass filter unit 53, and the second high-pass filter unit 54 according to the present embodiment, the input music signal L is converted into the low-frequency music signal L. The filter characteristics for dividing the band into the mid-range music signal L and the high-range music signal L are as shown in FIG. Specifically, the cutoff frequency of the first low-pass filter unit 51 is 300 Hz, the cutoff frequency of the second low-pass filter unit 52 is 6,000 Hz, the cutoff frequency of the first high-pass filter unit 53 is 300 Hz, and the second high-pass filter unit. The cutoff frequency of 54 is set to 6,000 Hz.

  The first low-pass filter unit 51, the second low-pass filter unit 52, the first high-pass filter unit 53, and the second high-pass filter unit 54 in which the cutoff frequency is set in this way are combined as shown in FIG. Thus, the music signal L input to the first band dividing unit 4 is divided into a low-frequency music signal (non-speech band signal) L subjected to band division as a low-frequency music signal and a mid-range music signal. Dividing into a mid-range music signal (first audio band signal) L subjected to band division and a high-frequency music signal (non-sound band signal) L subjected to band division as a high-frequency music signal. It becomes possible. Here, the bandwidth of the mid-range music signal is a band corresponding to the audio band of the audio signal, as is clear from the cutoff frequency of the second low-pass filter unit 52 and the cutoff frequency of the first high-pass filter unit 53. Is set.

  The phase inverting unit 55 has a role of inverting the phase of the mid-range music signal L that has been filtered by the second low-pass filter unit 52 and the first high-pass filter unit 53. In the case of generating a mid-range music signal by band division, as shown in FIG. 7, the second low-pass filter unit 52 is applied to the music signal and only the signal level with a cutoff frequency of 6,000 Hz or less is music. Only the signal level with a cut-off frequency of 300 Hz or higher is recognized as the output of the music signal by applying the first high-pass filter unit 53. For this reason, by inverting the phase of the mid-range music signal by the phase inverting unit 55, the output values of the low-frequency part and the high-frequency part are emphasized, and the band is divided by the first synthesizing unit 7. In the case of synthesizing each music signal, it is possible to reduce interference generated between the respective signals.

  The second band dividing unit 5 has substantially the same configuration as that of the first band dividing unit 4, and the input music signal R is filtered and processed by a low frequency music signal R, a middle frequency music signal R, and a high frequency band. Band division into music signal R is performed. Since the configuration of the second band dividing unit 5 is the same as that of the first band dividing unit 4, a detailed description thereof is omitted here.

[Volume control section]
Next, the volume control unit 6 will be described. FIG. 8 is a block diagram showing a schematic configuration of the volume control unit 6. As shown in FIG. 8, the volume control unit 6 includes a phase inversion unit (phase inversion unit) 61, a synthesis unit (audio band signal synthesis unit) 62, a first gain unit (gain addition unit) 63, and a second unit. A gain section (gain addition means) 64 and a selection section (voice band signal selection means) 65 are provided.

  The phase inversion unit 61 has a role of performing phase inversion on the mid-range music signal R band-divided by the second band-splitting unit 5, and the synthesizing unit 62 has the phase-inverted mid-range music signal R And the middle band music signal L which has been band-divided by the first band dividing unit 4. In this way, by synthesizing the mid-range music signal R whose phase has been inverted with respect to the mid-range music signal L, a monaural component that is a common music component between the mid-range music signal R and the mid-range music signal L is obtained. Will be removed.

  In particular, the mid-range music signal R and the mid-range music signal L correspond to the voice band component in the music signal, that is, the vocal band in the music signal. Therefore, the monaural component in the vocal band is removed by the synthesis processing by the synthesis unit 62. Will be done. On the other hand, in the synthesized mid-range music signal (second audio band signal), a music component that is not common to the mid-range music signal R and the mid-range music signal L is left. The stereo component is retained in the music signal.

  The first gain unit 63 and the second gain unit 64 perform phase and level correction by weighting the L channel and R channel gains to the mid-range music signal from which the monaural component of the vocal band has been removed. Has a role to do. The mid-range music signal from which the monaural component of the vocal band has been removed (the stereo component is retained) is weighted with a gain of 0.707 (1 / √2) in the first gain unit 63, and In the 2-gain unit 64, gain weighting of -0.707 (-1 / √2) times is performed. The mid-range music signal weighted by the first gain unit 63 is an L-channel mid-range music signal, and the mid-range music signal weighted by the second gain unit 64 is the R-channel mid-range music signal. It becomes a music signal.

  As a result of this weighting process, the mid-range music signals for the L channel and the R channel become uncorrelated signals between the channels, and the respective stereo components are emphasized. When the music signal of each band is synthesized in the synthesis unit 7 and the second synthesis unit 8, it is possible to prevent the music signal from interfering with the stereo component, and to reduce the deterioration of the sound quality of the stereo component. It becomes possible.

  The selection unit 65 outputs the mid-range music signal L ′ output from the volume control unit 6 to the first synthesis unit 7 and the output to the second synthesis unit 8 in accordance with the voice detection signal output from the voice detection unit 3. It has the role of making a determination with the regional music signal R ′. Specifically, when the selection unit 65 determines that the voice is detected based on the voice detection signal, the monaural component is removed as the mid-range music signal L ′ to be output to the first synthesis unit 7. The mid-range music signal for the L channel from which the monaural component has been removed is selected as the mid-range music signal R ′ to be output to the second synthesis unit 8 while the mid-range music signal for the L channel is selected.

  On the other hand, when the selection unit 65 determines that no voice is detected based on the voice detection signal, the monaural component is removed as the mid-range music signal L ′ output to the first synthesis unit 7. A mid-range music signal L ′ for the L channel that is not yet selected (that is, the mid-range music signal L band-divided by the first band dividing unit 4) and output to the second synthesizing unit 8 is selected. The R channel mid-band music signal R from which the monaural component has not been removed (that is, the mid-band music signal R band-divided by the second band dividing unit 5) is selected. The mid-range music signal for L channel (mid-range music signal L ′) selected by the selection unit 65 is output to the first synthesizing unit 7, and the selected mid-range music signal (mid-range music signal for R channel). R ′) is output to the second synthesis unit 8.

[First synthesis unit and second synthesis unit]
The first synthesizing unit 7 synthesizes the low-frequency music signal L and the high-frequency music signal L band-divided by the first band dividing unit 4 with the mid-range music signal L ′ output from the volume control unit 6. It has a role to output to the first power amplifier 9. When the audio signal is acquired by the microphone M, the mid-range music signal L ′ (corresponding to the audio band) synthesized by the first synthesis unit 7 is in a state where the monaural component is removed. For this reason, when a conversation is performed in the passenger compartment, only the monaural component of the middle region that can hinder the conversation is reduced, and a smooth conversation can be performed. Furthermore, when an audio signal is acquired by the microphone M, the monaural component of the music is removed, but the stereo component is not removed, so that the sound quality of the music (such as presence and power) is impaired. There is no.

  Similarly to the first synthesizing unit 7, the second synthesizing unit 8 also includes the low-frequency music signal R and the high-frequency music signal R that are band-divided by the second band dividing unit 5, and the midrange output by the volume control unit 6 The music signal R ′ is combined and output to the second power amplifier 10. Also in the second synthesizing unit 8, when the audio signal is acquired by the microphone M, the monaural component of the midrange music signal corresponding to the audio band is removed, so that the monaural component of the midband component that can hinder the conversation Only the components are reduced, and a smooth conversation can be performed. Furthermore, since the stereo component is not removed, the sound quality of the music (such as presence and power) is not impaired.

  The first power amplifier 9 amplifies the music signal output from the first synthesis unit 7 and then outputs it from the speaker 11a. The second power amplifier 10 amplifies the music signal output from the second synthesis unit 8 after amplification. Output from the speaker 11b.

  Next, the frequency characteristics of the music signal when the monaural component of the music signal is reduced and when it is not reduced will be described with reference to FIGS. 9 to 12 using the automatic volume control device 1 described above. To do.

  FIG. 9 shows an example of frequency characteristics of an audio signal acquired by the microphone M. In the case where such an audio signal is acquired by the microphone M, a case where white noise as shown in FIGS. 10 and 11 is output from a sound source as an example of a music signal will be described. The white noise shown in FIG. 10 shows a case where the white noise for the L channel and the white noise for the R channel are the same monaural signal, and the white noise shown in FIG. 11 is the same as the white noise for the L channel. A case is shown in which the white noise for the R channel is an uncorrelated stereo signal.

  In the state where the volume control is not performed as shown in FIG. 10, the frequency characteristic peculiar to white noise (the signal level is constant regardless of the frequency) is shown, but in the state where the volume control is performed, It is shown that the signal level in the middle band is significantly reduced. In the case of the white noise shown in FIG. 10, since the white noise for the L channel and the white noise for the R channel are the same monaural signal and there are many monaural components in the middle band, the signal level is significantly reduced. Will appear. By reducing the signal level of the monaural component in the middle band, smooth conversation can be realized.

  On the other hand, in the case of FIG. 11, in a state where the volume control is not performed, a frequency characteristic peculiar to white noise (a signal level is constant regardless of the frequency) is shown, but in a state where the volume control is performed. Even in such a case, the frequency characteristic peculiar to white noise (the signal level is constant regardless of the frequency) is shown, and it is difficult to determine a change in the signal level depending on the presence or absence of control. In the case of the white noise shown in FIG. 11, since the white noise for the L channel and the white noise for the R channel are uncorrelated stereo signals, there is no monaural component in the middle band, and the signal level This is because there is no reduction. As described above, since the signal level in the middle band is not reduced for stereo components other than the monaural component, it is possible to maintain the sound quality of the music (such as presence and power).

  FIG. 12 shows frequency characteristics when a music signal including a vocal sound is used. In the music signal in FIG. 12, the music signals for the L channel and the R channel are stereo components (including monaural components). It shall be equipped with. As shown in FIG. 12, when comparing the frequency characteristics of music signals when the volume control is performed and when the volume control is not performed, only the signal level of the middle band corresponding to the vocal band (voice band) is obtained. The signal levels of the high band and the low band are largely attenuated, and are almost in the same state regardless of whether or not the volume control is performed.

  In the case shown in FIG. 12, only the middle band monaural component corresponding to the vocal band is reduced, so that the signal level of the middle band is reduced. However, the monaural component reduction processing is performed for the high band and the low band. Therefore, the signal level does not change. On the other hand, regarding the stereo component of the music signal, since the signal level does not decrease regardless of whether or not the volume control is performed, the output is ensured uniformly at almost all frequencies.

  As described above, in the stereo signal that is an uncorrelated signal between the L channel and the R channel, interference occurs even if the first synthesis unit 7 and the second synthesis unit 8 perform the music signal synthesis process. Absent. On the other hand, with respect to the monaural component in the middle band (vocal band), as shown in FIG. 12, the signal level is attenuated by about 10 dB, and the voice band (vocal band, middle band) that can hinder the conversation. Only the signal level (volume) can be effectively reduced (controlled). Note that the amount of attenuation of the signal level in the middle band varies according to the vocal level of music, and the attenuation increases relatively as the vocal level increases.

  As described above, according to the automatic volume control device 1 according to the present embodiment, the music signal L for the L channel is banded into the low frequency music signal L, the mid frequency music signal L, and the high frequency music signal L. Similarly, after dividing the music signal R for the R channel into the low-frequency music signal R, the mid-range music signal R, and the high-frequency music signal R, the phase of the mid-range music signal R for the R channel is divided. By inverting and synthesizing with the mid-range music signal L for the L channel, it is possible to reduce the output (signal level) of the monaural component in the audio band (mid-band) of the music signal output from the speakers 11a and 11b.

  For this reason, it is possible to effectively reduce only the signal level of the monaural component in the voice band that can hinder conversation while maintaining the overall stereo feeling of the music signals output from the speakers 11a and 11b. It is possible to realize a smooth conversation environment even in a situation where the current flows. In addition, the overall stereo feeling of the music signal (especially the stereo components of the low-frequency music signal and high-frequency music signal) can be maintained even during a conversation, so the sound quality of the music can be achieved while realizing a smooth conversation. It is possible to suppress a decrease in (existence, force, etc.).

  In particular, in the automatic volume control device 1 according to the present embodiment, a weighting process is performed for each channel on the mid-range music signal whose monaural component has been reduced in the first gain unit 63 and the second gain unit 64. Therefore, even if the monaural component is reduced in the mid-range music signal, the remaining stereo component is emphasized, and the overall stereo feeling including the voice band (vocal band) can be maintained high. It becomes possible.

  Further, since the first gain unit 63 and the second gain unit 64 perform weighting processing on the mid-range music signal, the L-channel mid-range music signal and the R-channel mid-range music signal become uncorrelated signals. In the case where the music signals are synthesized in the first synthesis unit 7 and the second synthesis unit 8, it is possible to prevent the stereo component music signals from interfering with each other.

  Furthermore, since the output reduction of the monaural component in the voice band described above is controlled by the volume control unit 6 when the voice detection unit 3 determines that a voice signal has been detected by the microphone M, a conversation by the speaker is performed. The music signal level can be automatically reduced only when In addition, since the volume control of the music signal is performed according to the voice detection in this way, the volume reduction control of the monaural component in the voice band is not performed in a state where the conversation is not performed (a state where the conversation is interrupted). Therefore, when there is no conversation (when no audio signal is detected), the signal level reduction processing (volume control) of the music signal is not performed at all, and it is possible to enjoy high-quality music. It becomes.

  Further, since the phase inversion is performed in advance in the phase inversion unit 55 for the mid-range music signal that has been subjected to the band division in the first band division unit 4 and the second band division unit 5, When the two synthesis unit 8 synthesizes the audio signal together with the high frequency music signal and the low frequency music signal, it is possible to reduce the occurrence of interference between the signals.

  Further, by applying the LMS adaptive algorithm in the first adaptive filter unit 25 and the second adaptive filter unit 26 of the audio canceller unit 15, signal components (noise components) other than the audio signal component in the audio signal are effectively reduced. Furthermore, since the noise signal is removed by the noise canceller 16, it is possible to improve the detection accuracy of the audio signal in the audio detector 3.

  The automatic volume control device according to the present invention described above in detail with reference to the drawings for the automatic volume control device according to the present invention is not limited to the above-described embodiment. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  For example, in the present embodiment, the case where the automatic volume control device 1 is installed in a vehicle is shown as an example. However, the automatic volume control device according to the present invention is not limited to that installed in a vehicle, such as music. As long as it is an environment in which a conversation may be hindered by music or the like in a space where music flows, the present invention is not limited to a vehicle interior, and can be installed in a different environment / space.

  In the present embodiment, the music signal L for the L channel is divided into three bands, ie, the low frequency music signal L, the mid frequency music signal L, and the high frequency music signal L, and similarly the music for the R channel. Although the case where the signal R is band-divided into three bands of the low-frequency music signal R, the mid-range music signal R, and the high-frequency music signal R has been described, the automatic volume control device according to the present invention is divided into three bands. It is not limited to what is done. In the automatic sound volume control device according to the present invention, it is only necessary to be able to at least divide a band into a voice band signal including a voice band and a non-voice band signal including a band other than the voice band. However, the present invention is not limited to the high frequency signal and the low frequency signal, and may be further divided into bands.

DESCRIPTION OF SYMBOLS 1 ... Automatic volume control apparatus 2 ... Voice emphasis processing part 3 ... Voice detection part (voice detection means)
4 ... 1st band division part (band division means)
5: Second band dividing unit (band dividing means)
6 ... Volume control section (phase inversion means, voice band signal synthesis means, voice band signal selection means, gain addition means)
7 ... 1st synthetic | combination part (music signal synthetic | combination means)
8 ... 2nd synthetic | combination part (music signal synthetic | combination means)
9 ... 1st power amplifier 10 ... 2nd power amplifier 11a, 11b ... Speaker 15 ... Audio canceller part 16 (of speech enhancement processing part) ... Noise canceller part 21 (of speech enhancement processing part) ... 1st (of audio canceller part) Bandpass filter unit 22 ... second bandpass filter unit 23 (of the audio canceller unit) ... first delay unit 24 (of the audio canceller unit) ... second delay unit 25 (of the audio canceller unit) ... (of the audio canceller unit) First adaptive filter section 26 ... second adaptive filter section 28 (of the audio canceller section) ... first FIR section 29 (of the first adaptive filter section) ... first LMS section 30 (of the first adaptive filter section) (first adaptation) First addition unit 31 (of the filter unit) ... Second FIR unit 32 (of the second adaptive filter unit) ... (Second adaptive filter) Second LMS unit 33 (second adaptive filter unit) second addition unit 41 (speech detection unit) effective value detection unit 42 (speech detection unit) moving average unit 43 (sound detection unit) ) Audio detection threshold unit 44 ... Audio signal hold unit 51 (of the audio detection unit) ... First low-pass filter unit 52 (of the first band dividing unit) ... Second low-pass filter unit 53 (of the first band dividing unit) ( First high-pass filter unit 54 (of the first band division unit) ... Second high-pass filter unit 55 (of the first band division unit) ... Phase inversion unit 61 (of the first band division unit) ... Phase inversion (of the volume control unit) Part (phase inversion means)
62 ... Synthesizer (of sound volume controller) (voice band signal synthesizer)
63... First gain section (gain adding means) (of volume control section)
64 ... 2nd gain part (of gain control part) (gain addition means)
65 ... Selection unit (of sound volume control unit) (voice band signal selection means)
M ... Mike

Claims (3)

Voice detection means for detecting a voice signal from an acoustic signal acquired by a microphone;
Each of the L channel music signal and the R channel music signal output by the sound source includes a first voice band signal having a frequency band corresponding to a voice band and a non-frequency band having a frequency band not corresponding to the voice band. Band dividing means for dividing a band into audio band signals;
Phase inversion means for performing phase inversion of only one audio band signal of the first audio band signal for the L channel and the first audio band signal for the R channel;
Either the first audio band signal for the L channel or the first audio band signal for the R channel that has been phase-inverted by the phase inversion means, and the phase inversion has not been inverted by the phase inversion means. Voice band signal synthesis means for generating a second voice band signal for the L channel and a second voice band signal for the R channel by synthesizing the other voice band signal;
Phase correction means for performing phase inversion correction on the other of the second audio band signal for L channel and the second audio band signal for R channel generated by the audio band signal synthesis means;
If the detection of the audio signal has been performed by the voice detection means, the first for the second voice band signal and R channel for L channel is one of the phase inversion correction has been performed by the phase correcting means When two audio band signals are output as audio band signals for each channel and the audio signal is not detected by the audio detecting means, the second channel for the L channel divided by the band dividing means is used. Audio band signal selection means for outputting one audio band signal and the first audio band signal for the R channel as they are as an audio band signal for each channel;
The voice band signal for L channel and the voice band signal for R channel output by the voice band signal selecting means, the non-voice band signal for L channel and the non-voice for R channel divided by the band dividing means. And a music signal synthesizing unit for synthesizing the audio band signal for each channel to generate a full-band music signal for the L channel and a full-band music signal for the R channel. Automatic volume control device. The phase correction means has a gain corresponding to each channel for the second audio band signal for L channel and the second audio band signal for R channel synthesized by the audio band signal synthesis means. The automatic volume control device according to claim 1, further comprising a function as a gain addition unit that performs weighting of the sound. The band dividing unit is configured to convert each of the music signal for L channel and the music signal for R channel output from the sound source into the first audio band signal having a frequency band corresponding to an audio band, The frequency band is divided into a low frequency signal consisting of a frequency band lower than the audio band and a high frequency signal consisting of a frequency band higher than the audio band, and only the first audio band signal is phase-inverted,
The music signal synthesizing unit is configured to output the audio band signal for the L channel and the audio band signal for the R channel output from the audio band signal selecting unit, the low frequency signal for the L channel, and the audio signal for the R channel. By combining the low frequency signal, the high frequency signal for the L channel and the high frequency signal for the R channel for each channel, the music signal for the entire band for the L channel and the total for the R channel The automatic sound volume control device according to claim 1 or 2, wherein a band music signal is generated.

Images