JP2007110451A - Speech signal adjustment apparatus, speech signal adjustment method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a speech signal adjustment apparatus or the like for simply adjusting sound quality at a high speed or accurately. <P>SOLUTION: A sub-band data input section 1 acquires a sub-band data group representing a temporal change in the strength of a fundamental frequency component or harmonics of speech, and on the other hand, a calibration data generating section 3 receives the speech signals reproduced by an speech signal reproducing section 5 to generate a calibration sub-band data group representing the speech signals. A speech quality adjustment section 4 adjusts the strength of each sub-band data in the sub-band data group on the basis of the calibration sub-band data group or according to speech quality designation data when a speech quality designation data input section acquires the speech quality designation data. The speech signal reproducing section 5 reproduces the speech represented by the sub-band data group after the adjustment. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an audio signal adjustment device, an audio signal adjustment method, and a program.

  When playing back sound using sound data, in general, the difference between the original sound represented by the sound data and the actually reproduced sound is corrected, noise is removed from the reproduced sound, or the original sound is audibly heard. The sound quality of the reproduced sound is adjusted for the purpose of adding the above special effect or the like.

Conventionally, the sound quality is adjusted by causing a sound reproduction device equipped with an equalizer to reproduce sound using test sound data, receiving the reproduced sound, and receiving the received sound waveform and test sound data. The frequency characteristic of the equalizer is determined based on the difference from the waveform represented by, and the equalizer is operated so as to obtain the determined frequency characteristic (see, for example, Patent Document 1). As the test voice data, for example, data representing an impulse waveform or a sweep waveform has been used.
JP 2001-197585 A

  However, the conventional equalizer has a complicated configuration and a high manufacturing cost. In addition, an equalizer that changes the frequency characteristic accurately so as to be determined has a complicated structure, and it is difficult to manufacture this equalizer both technically and economically. In addition, when an impulse waveform is used as a test sound, the reproduced sound band becomes extremely wide, so that it is difficult to accurately specify the frequency characteristic, and therefore, the determination result of the frequency characteristic of the equalizer tends to be inappropriate. Further, when a sweep waveform is used as a test sound, it takes a long time to specify the frequency characteristics of the reproduced sound.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide an audio signal adjustment device, an audio signal adjustment method, and a program for adjusting sound quality simply, at high speed or accurately.

In order to achieve the above object, an audio signal adjustment device according to a first aspect of the present invention is provided.
Audio signal acquisition means for acquiring an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of audio from the outside;
An audio signal adjustment unit that changes the intensity of an audio signal included in the audio signal group acquired by the audio signal acquisition unit;
Waveform generating means for generating a signal representing a waveform of a voice represented by the voice signal group based on the voice signal group whose intensity of the voice signal is changed,
It is characterized by that.

The audio signal adjustment device may further include designation data acquisition means for acquiring from the outside specification data for specifying a mode of changing the intensity of the audio signal included in the audio signal group acquired by the audio signal acquisition means.
In this case, the audio signal adjusting unit changes the intensity of the audio signal included in the audio signal group acquired by the audio signal acquiring unit in a manner specified by the specified data acquired by the specified data acquiring unit. May be.

The audio signal adjusting device receives a sound, and includes a calibration audio signal group including a calibration audio signal representing a temporal change in intensity of a fundamental frequency component and a harmonic component of an audio signal to be processed that represents a waveform of the audio. May further comprise a calibration audio signal generating means for generating
In this case, the audio signal adjustment unit substantially changes the intensity of the audio signal included in the audio signal group acquired by the audio signal acquisition unit to the intensity of the audio signal and the audio signal. It may be determined based on the intensity of the calibration audio signal representing the component of the same frequency, and the intensity of the audio signal may be changed according to the determination result.

The calibration audio signal generation means includes:
Means for generating a signal representing the waveform of the received sound and processing the signal into a pitch waveform signal by making the time lengths of the sections corresponding to the unit pitch of the signal substantially the same;
And means for generating a signal representing a temporal change in intensity of the fundamental frequency component and the harmonic component of the pitch waveform signal as the calibration audio signal.

An audio signal adjustment method according to the second aspect of the present invention is as follows.
Obtain an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of the audio from the outside,
Change the intensity of the audio signal included in the acquired audio signal group,
Based on the audio signal group in which the intensity of the audio signal is changed, a signal representing the waveform of the audio represented by the audio signal group is generated.
It is characterized by that.

A program according to the third aspect of the present invention is:
Computer
Audio signal acquisition means for acquiring an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of audio from the outside;
An audio signal adjustment unit that changes the intensity of an audio signal included in the audio signal group acquired by the audio signal acquisition unit;
A waveform generating means for generating a signal representing a waveform of a voice represented by the voice signal group based on the voice signal group in which the intensity of the voice signal is changed;
It is for making it function.

  According to the present invention, an audio signal adjustment device, an audio signal adjustment method, and a program for adjusting sound quality simply, quickly, or accurately are realized.

Hereinafter, embodiments of the present invention will be described with reference to the drawings, taking a sound quality adjusting device as an example.
FIG. 1 is a diagram showing a configuration of the sound quality adjusting device. As shown in the figure, this sound quality adjusting device is composed of a subband data input unit 1, a sound quality designation data input unit 2, a calibration data generation unit 3, a sound quality adjustment unit 4, and an audio playback unit 5. Yes.

  The subband data input unit 1 is, for example, a recording medium driver (flexible disk drive, MO drive, etc.) for reading data recorded on a recording medium (for example, a flexible disk or an MO (Magneto Optical disk)), or It consists of a USB (Universal Serial Bus) interface circuit and the like, and a communication control device that controls data exchange with the outside.

  The subband data input unit 1 acquires a subband data group representing the sound and supplies it to the sound quality adjustment unit 4. The subband data group includes the 0th subband data representing the time variation of the intensity of the fundamental frequency component of the sound and the first to the time variation of the intensity of n harmonic components (n is a natural number) of this sound. Data including n subband data up to the nth. Each subband data represents the intensity of the fundamental frequency component (or harmonic component) in the form of a DC signal when there is no temporal change in the intensity of the fundamental frequency component (or harmonic component) of the sound.

  In addition, when the sound represented by the subband data group is one in which the phase of each section is aligned by shifting each section corresponding to the unit pitch, the subband data input unit 1 If the pitch information about the sound represented by the data group can be acquired, this pitch information is also acquired and supplied to the sound reproducing unit 5. The pitch information is information representing the original value of the length (pitch length) of each section of the voice represented by the subband data group.

  The sound quality designation data input unit 2 includes, for example, an input device such as a keyboard and a pointing device, and a processor such as a CPU (Digital Signal Processor). The sound quality designation data input unit 2 acquires sound quality designation data according to this operation when an operation for inputting the sound quality designation data is performed by the operator. Then, the acquired sound quality designation data is supplied to the sound quality adjustment unit 4.

  The sound quality designation data is data for designating how the intensity of each subband data constituting the subband data group acquired by the subband data input unit 1 should be changed. For example, each subband data is It consists of data representing a coefficient to be multiplied by the intensity of the component to be represented.

  The calibration data generation unit 3 includes a calibration voice input unit 31, a pitch extraction unit 32, and a subband analysis unit 33.

  The calibration audio input unit 31 includes a sound receiving device including a microphone, an AF (Audio Frequency) amplifier, a sampler, an A / D (Analog-to-Digital) converter, a PCM encoder, and the like. The calibration voice input unit 31 amplifies a voice signal representing voice received by its own microphone, performs sampling and A / D conversion, and then generates calibration voice data representing the sampled voice signal. This is supplied to the pitch extraction unit 32.

  Note that the calibration voice data only needs to have, for example, a PCM (Pulse Code Modulation) modulated digital signal format, and the voice received by the calibration voice input unit 31 is a constant sufficiently shorter than the pitch. It suffices to represent the result of sampling at a period of.

  Each of the pitch extraction unit 32 and the subband analysis unit 33 includes a processor such as a DSP (Digital Signal Processor) or a CPU and a memory such as a RAM (Random Access Memory). A single processor or a single memory may perform a part or all of the functions of the pitch extraction unit 32 and the subband analysis unit 33. Further, a processor that performs the function of the sound quality designation data input unit 2 may perform a part or all of the functions of the pitch extraction unit 32 and the subband analysis unit 33 in common.

  The function of the pitch extraction unit 32 is, for example, as shown in FIG. 2, a cepstrum analysis unit 321, an autocorrelation analysis unit 322, a weight calculation unit 323, and a BPF (Band Pass Filter) coefficient calculation. A unit 324, a band pass filter 325, a zero cross analysis unit 326, a waveform correlation analysis unit 327, a phase adjustment unit 328, and a resampling unit 329 are configured.

  A single processor or a single memory includes a cepstrum analysis unit 321, an autocorrelation analysis unit 322, a weight calculation unit 323, a BPF (Band Pass Filter) coefficient calculation unit 324, a band pass filter 325, a zero cross analysis unit 326, a waveform. A part or all of the functions of the correlation analysis unit 327, the phase adjustment unit 328, and the resampling unit 329 may be performed.

  The cepstrum analysis unit 321 performs cepstrum analysis on the calibration voice data supplied from the calibration voice input unit 31 to identify the fundamental frequency and formant frequency of the voice represented by the calibration voice data. Then, data indicating the specified fundamental frequency is generated and supplied to the weight calculation unit 323, and data indicating the specified formant frequency is generated and supplied to the subband analysis unit 33.

  Specifically, when the cepstrum analysis unit 321 is supplied with the calibration voice data from the calibration voice input unit 31, first, the spectrum of the calibration voice data is converted to a fast Fourier transform technique (or a discrete variable). Is obtained by any other method for generating data representing the result of Fourier transform.

Next, the cepstrum analysis unit 321 converts the intensity of each component of the obtained spectrum into a value corresponding to the logarithm of each original value. (The base of the logarithm is arbitrary, and may be a common logarithm, for example.)
Next, the cepstrum analysis unit 321 represents the result of performing inverse Fourier transform on the spectrum whose value has been converted (that is, the cepstrum) and the fast inverse Fourier transform method (or the result of inverse Fourier transform of a discrete variable). Any other method for generating data).

Then, based on the obtained cepstrum, the cepstrum analysis unit 321 identifies the fundamental frequency of the voice represented by the cepstrum, generates data indicating the identified fundamental frequency, and supplies the data to the weight calculation unit 323.
Specifically, the cepstrum analysis unit 321 extracts a frequency component (long component) equal to or higher than a predetermined quefrency from the cepstrum by filtering (ie, lifting) the obtained cepstrum, for example. The fundamental frequency may be specified based on the position of the component peak.

  When the autocorrelation analysis unit 322 is supplied with the calibration audio data from the calibration audio input unit 31, the autocorrelation analysis unit 322 performs an analysis based on the autocorrelation function of the waveform of the calibration audio data, thereby expressing the audio represented by the calibration audio data. The basic frequency is specified, data indicating the specified basic frequency is generated and supplied to the weight calculation unit 323.

  Specifically, when the autocorrelation analysis unit 322 is supplied with the calibration audio data from the calibration audio input unit 31, first, the autocorrelation function r (l) specified by the right side of Equation 1 is specified.

  Next, the autocorrelation analysis unit 322 calculates a minimum value exceeding a predetermined lower limit value as a fundamental frequency among frequencies giving a maximum value of a function (periodogram) obtained as a result of Fourier transform of the autocorrelation function r (l). Is generated, and data indicating the specified fundamental frequency is generated and supplied to the weight calculation unit 323.

  When the weight calculation unit 323 is supplied with a total of two pieces of data indicating the fundamental frequency one by one from the cepstrum analysis unit 321 and the autocorrelation analysis unit 322, the average of the absolute values of the reciprocals of the fundamental frequencies indicated by these two data items Ask for. Then, data indicating the obtained value (that is, average pitch length) is generated and supplied to the BPF coefficient calculation unit 324.

  When the BPF coefficient calculation unit 324 is supplied with data indicating the average pitch length from the weight calculation unit 323 and is supplied with a zero cross signal described later from the zero cross analysis unit 326, the average pitch length is based on the supplied data and the zero cross signal. It is determined whether or not the pitch signal and the zero-crossing period differ from each other by a predetermined amount or more. When it is determined that they are not different, the frequency characteristic of the bandpass filter 325 is controlled so that the reciprocal of the zero-crossing period is the center frequency (the center frequency of the passband of the bandpass filter 325). On the other hand, when it is determined that they are different by a predetermined amount or more, the frequency characteristic of the bandpass filter 325 is controlled so that the reciprocal of the average pitch length is set as the center frequency.

The band pass filter 325 performs a function of a FIR (Finite Impulse Response) type filter having a variable center frequency.
Specifically, the bandpass filter 325 sets its center frequency to a value according to the control of the BPF coefficient calculation unit 324. Then, the calibration voice data supplied from the calibration voice input unit 31 is filtered, and the filtered calibration voice data (pitch signal) is supplied to the zero cross analysis unit 326 and the waveform correlation analysis unit 327. The pitch signal is assumed to be digital data having a sampling interval substantially the same as the sampling interval of the calibration audio data. Note that the bandwidth of the bandpass filter 325 is desirably such that the upper limit of the passband of the bandpass filter 325 always falls within twice the fundamental frequency of the voice represented by the calibration voice data.

The zero-cross analysis unit 326 specifies the timing when the time when the instantaneous value of the pitch signal supplied from the band-pass filter 325 becomes 0 (time when zero-crossing) comes, and the signal representing the specified timing (zero-cross signal) is represented by the BPF coefficient. It supplies to the calculation part 324.
However, the zero cross analysis unit 326 specifies the timing at which the instantaneous value of the pitch signal becomes a predetermined value other than 0, and supplies a signal representing the specified timing to the BPF coefficient calculation unit 324 instead of the zero cross signal. You may make it do.

  When the waveform correlation analysis unit 327 is supplied with the calibration voice data from the calibration voice input unit 31, the calibration is performed at the timing when the boundary of the unit period (for example, one cycle) of the pitch signal supplied from the band pass filter 325 comes. Separate audio data. Then, for each of the divided sections, the correlation between the variously changed calibration audio data phases in this section and the pitch signal in this section is obtained, and the calibration voice when the correlation becomes the highest The phase of the data is specified as the phase of the calibration audio data in this interval.

  Specifically, the waveform correlation analysis unit 327, for each section, for example, varies the value cor represented by the right side of Equation 2 with various values of φ (where φ is an integer of 0 or more) representing the phase. Each change is obtained for each change. Then, the waveform correlation analysis unit 327 specifies the value ψ of φ that maximizes the value cor, generates data indicating the value ψ, and uses it as phase data representing the phase of the calibration audio data in this interval. This is supplied to the phase adjustment unit 328.

  Note that the time length of the section is preferably about one pitch. The longer the interval, the greater the number of samples in the interval and the amount of calibration pitch waveform data (described later) increases, or the sampling interval increases and the voice represented by the calibration pitch waveform data becomes inaccurate. The problem arises.

  When the phase adjustment unit 328 is supplied with the calibration voice data from the calibration voice input unit 31 and the waveform correlation analysis unit 327 is supplied with the data indicating the phase Ψ of each section of the calibration voice data, By shifting the phase of the calibration audio data by (−Ψ), the phases of the respective sections are made uniform. Then, the phase-shifted calibration audio data (calibration pitch waveform data) is supplied to the resampling unit 329.

  The resampling unit 329 resamples (resamples) each section of the calibration audio data supplied from the phase adjustment unit 328, and supplies the resampled calibration pitch waveform data to the subband analysis unit 33. .

  However, the resampling unit 329 performs resampling so that the number of samples in each section of the calibration audio data is a constant number that is substantially equal to each other, and is equally spaced within the same section. For a section where the number of samples is less than this fixed number, by adding a sample having a value that interpolates between adjacent samples on the time axis by a predetermined method (for example, Lagrangian interpolation), Align the number of samples to this fixed number.

The subband analysis unit 33 generates a calibration subband data group by performing orthogonal transformation such as DCT (Discrete Cosine Transform) on the calibration pitch waveform data supplied from the resampling unit 329, and generates the generated calibration. The subband data group is supplied to the sound quality adjustment unit 4.
The calibration subband data group includes zeroth calibration subband data representing the temporal change in intensity of the fundamental frequency component of the voice represented by the calibration pitch waveform data supplied to the subband analysis unit 33, and n of this voice. This is data including the first to n-th calibration subband data representing the time change of the intensity of the harmonic components (n is the above-mentioned natural number).

  When the sound quality adjustment unit 4 is supplied with the subband data group from the subband data input unit 1 and the subband data group for calibration is supplied from the subband analysis unit 33 of the calibration data generation unit 3, Each subband data in the subband data group is changed so that the intensity of the kth subband data in the data group (k is an integer of 0 or more and n or less) becomes the value Y (k) shown in Equation 3. . Then, the subband data group whose value has been changed is supplied to the subband synthesis unit 51.

(Equation 3)
Y (k) = {α · X (k)} ² / {R (k)}
(Where X (k) is the intensity before the change of the kth subband data in the subband data group, R (k) is the intensity of the kth subband data in the calibration subband data group, and α is Predetermined proportionality factor)

  When the calibration data generating unit 3 receives the sound reproduced by the sound reproducing unit 5, the sound represented by the subband data group supplied from the sound quality adjusting unit 4 to the sound reproducing unit 5 is the sound reproducing unit. 5, the time length from when the subband data group for calibration representing the sound is generated and supplied to the sound quality adjustment unit 4 is short enough to be ignored, and R (k) is Y (k). The value of Y (k) is substantially adjusted to a value proportional to {α · X (k)} under the condition that it can be regarded as proportional to. (However, it is assumed that the sound quality designation data is not supplied from the sound quality designation data input unit 2)

  However, when the sound quality designation data is supplied from the sound quality designation data input section 2, the sound quality adjustment section 4 determines the intensity of each subband data in the subband data group after changing the value, By further changing the intensity to the designated intensity, the sound quality of the voice represented by the subband data group as a whole is adjusted. For example, if the sound quality designation data represents a coefficient to be multiplied by the intensity of the component represented by the subband data, the subband data is set so that the product of the intensity of the component and the coefficient becomes a new intensity. Change the strength of the. Then, a subband data group that has been subjected to sound quality adjustment by further changing the intensity is supplied to the subband synthesizing unit 51.

The audio reproduction unit 5 includes a subband synthesis unit 51, an audio waveform restoration unit 52, and an audio output unit 53.
Of these, each of the subband synthesizing unit 51 and the speech waveform restoring unit 52 includes a processor such as a DSP or a CPU and a memory such as a RAM. A single processor or a single memory may perform a part or all of the functions of the subband synthesizing unit 51 and the speech waveform restoring unit 52. The processor that performs some or all of the functions of the sound quality designation data input unit 2, the pitch extraction unit 32, and the subband analysis unit 33 performs some or all of the functions of the subband synthesis unit 51 and the speech waveform restoration unit 52. You may make it carry out in common.

  When the subband data unit 51 is supplied with the subband data group from the sound quality adjustment unit 4, the subband data unit 51 converts the subband data group into a pitch that represents the intensity of each frequency component by the subband data group. Restore and restore waveform data (that is, audio data in which the phase of each section is aligned by shifting each section corresponding to the unit pitch of the audio) or audio data that has not undergone the process of aligning the phase of each section The pitch waveform data or voice data thus processed is supplied to the voice waveform restoration unit 52.

  The conversion performed by the subband synthesizing unit 51 on the subband data group is substantially inversely related to the conversion performed on the audio data in order to generate the subband data group acquired by the subband data input unit 1. It is a certain conversion. Therefore, for example, when this subband data group is generated by applying DCT to pitch waveform data, the subband synthesizing unit 51 may apply IDCT (Inverse DCT) to this subband data group. .

If the data supplied from the subband synthesis unit 51 is pitch waveform data, the speech waveform restoration unit 52 uses the pitch information supplied from the subband data input unit 1 as the time length of each section of the pitch waveform data. Change to the indicated time length. The time length of the section may be changed by changing the interval and / or the number of samples in the section, for example. Then, the voice waveform restoration unit 52 supplies the pitch waveform data in which the time length of each section is changed (that is, voice data representing the restored voice) to the voice output unit 53.
On the other hand, if the data supplied from the subband synthesizing unit 51 is audio data that has not undergone the process of aligning the phases of the sections, the audio waveform restoring unit 52 converts the audio data into audio data representing the restored audio. To the audio output unit 53.

The audio output unit 53 includes, for example, a control circuit that functions as a PCM decoder, a D / A (Digital-to-Analog) converter, an AF (Audio Frequency) amplifier, a speaker, and the like.
When the voice output unit 53 is supplied with voice data representing the restored voice from the voice waveform restoration unit 52, the voice output unit 53 demodulates the voice data, performs D / A conversion and amplification, and outputs the obtained analog signal. The sound is reproduced by driving the speaker.

  By performing the operation described above under the condition that the calibration data generating unit 3 receives the sound reproduced by the sound reproducing unit 5, the sound quality adjusting device adjusts the sound quality of the sound represented by the subband data. To do.

  The sound quality adjustment is performed by changing the intensity of the component represented by the subband data. On the other hand, the subband data is used unless the temporal change in the intensity of the fundamental frequency component or the harmonic component of the sound is particularly steep. Since it can be regarded as a DC signal, the configuration of the sound quality adjusting device can be simple and can be easily manufactured.

  In addition, the processing for changing the intensity of the subband data that can be regarded as a DC signal is different from filtering by a finite-order filter and can accurately obtain a desired characteristic, so that the sound quality is accurately adjusted.

  In addition, since the sound quality adjusting apparatus can use the sound reproduced by itself based on arbitrary subband data acquired from the outside as the test sound, the sound quality is adjusted using a predetermined test signal. Therefore, it is not necessary to spend time, and it is possible to adjust the sound quality while reproducing the sound originally desired to be reproduced.

The configuration of the sound quality adjusting device is not limited to the above.
For example, the sound quality adjusting apparatus does not necessarily include both the sound quality designation data input unit 2 and the calibration data generation unit 3. When the sound quality adjusting device does not include the calibration data generating unit 3 (or when the calibration subband data is not supplied from the calibration data generating unit 3), the sound quality adjusting unit 4 includes the subband data input unit 1 The supplied subband data group may be treated as a subband data group whose subband data value has been changed. Then, the intensity of each subband data in the subband data group may be immediately changed to the intensity designated by the sound quality designation data.

The subband data input unit 1 may acquire subband data from the outside via a communication line such as a telephone line, a dedicated line, a satellite line, or the like. In this case, the subband data input unit 1 only needs to include a communication control device such as a modem.
Similarly, the sound quality designation data input unit 2 may include a communication control device, and sound quality designation data may be acquired from the outside via a communication line.
Note that one recording medium drive device or communication control device may perform the functions of the subband data input unit 1 and the sound quality designation data input unit 2 together.

In addition, the pitch extraction unit 32 may not include the cepstrum analysis unit 321 (or autocorrelation analysis unit 322). In this case, the weight calculation unit 323 has the cepstrum analysis unit 321 (or autocorrelation analysis unit 322). The reciprocal of the obtained fundamental frequency may be handled as the average pitch length as it is.
Further, the waveform correlation analysis unit 327 may supply the pitch signal supplied from the band pass filter 325 to the cepstrum analysis unit 321 as it is as a zero cross signal.

  Further, the sound quality adjustment unit 4 may remove noise from the subband data by filtering the subband data and substantially removing the AC component.

Although the embodiment of the present invention has been described above, the audio signal adjusting apparatus according to the present invention can be realized using a normal computer system, not a dedicated system.
For example, a personal computer including a microphone, a sampling circuit, an A / D converter, a D / A converter, a speaker, etc. is connected to the above-described subband data input unit 1, sound quality designation data input unit 2, calibration data generation unit 3, A sound quality adjusting apparatus that performs the above-described processing is configured by installing the program from a medium (CD-ROM, flexible disk, or the like) that stores the program for executing the operations of the sound quality adjusting unit 4 and the sound reproducing unit 5 can do.

  The above-described personal computer that executes this program performs the processing shown in FIGS. 3 to 4 as processing corresponding to the operation of the sound adjustment device of FIG. 3 to 4 are flowcharts showing processing executed by the personal computer.

  That is, first, the personal computer acquires the above-described subband data group from the outside (step S101 in FIG. 3), and the voice represented by the subband data group moves through each section corresponding to the unit pitch. If the phase information of the sections is aligned and the pitch information about the voice represented by the subband data group can be acquired, the pitch information is also acquired (step S101). In addition, if the operator performs an operation for inputting the sound quality designation data, the personal computer acquires the sound quality designation data according to this operation (step S101).

  On the other hand, the personal computer receives and samples the voice, samples and performs A / D conversion, thereby generating digital-format calibration voice data (step S102). Then, the calibration voice data is filtered to generate filtered calibration voice data (pitch signal) (step S103).

  In addition, this personal computer performs feedback processing based on the pitch length described later and the time when the instantaneous value of the pitch signal becomes zero (the time when zero crossing is performed), as the characteristics of filtering performed to generate the pitch signal. decide.

  That is, the personal computer specifies the fundamental frequency of the voice represented by the voice data by performing, for example, the cepstrum analysis described above or the analysis based on the autocorrelation function described above on the voice data generated by receiving the sound. Then, the absolute value (that is, the pitch length) of the reciprocal of the fundamental frequency is obtained (step S104). (Alternatively, this personal computer specifies two fundamental frequencies by performing both cepstrum analysis and analysis based on an autocorrelation function, and calculates the average of the absolute values of the reciprocals of these two fundamental frequencies as the pitch length. It may be.)

  On the other hand, this personal computer specifies the timing when the time at which the pitch signal crosses zero (step S105). The personal computer determines whether the pitch length and the zero crossing period of the pitch signal are different from each other by a predetermined amount or more (step S106). If it is determined that they are not different, the reciprocal of the zero crossing period is set. It is assumed that the above-described filtering is performed with the characteristics of the bandpass filter that sets the center frequency (step S107). On the other hand, if it is determined that they differ by a predetermined amount or more, the above-described filtering is performed with the characteristics of the bandpass filter such that the reciprocal of the pitch length is the center frequency (step S108).

  Next, the personal computer divides the audio data read from the recording medium at the timing when the unit period boundary of the generated pitch signal comes (specifically, the timing at which the pitch signal crosses zero) (step S109). Then, for each of the sections that can be divided, the correlation between the variously changed phases of the audio data in this section and the pitch signal in this section is obtained, and the phase of the audio data when the correlation becomes the highest is obtained. The phase of the audio data in this section is specified (FIG. 4, step S110). Then, the pitch waveform data for calibration is generated by shifting the respective sections of the audio data so as to have substantially the same phase (step S111). Specifically, the personal computer specifies the above-described value Ψ in step S110 for each section, and shifts the audio data in the section by (−Ψ) in step S111.

  Next, the personal computer resamples each section of the calibration pitch waveform data (step S112). This personal computer may be resampled so that the number of samples in each section of the pitch waveform data is substantially equal to each other, and is equally spaced within the same section. For the interval where the number of samples is less than this fixed number, the number of samples in this interval is set to this fixed number by adding a sample having a value that interpolates between adjacent samples on the time axis by a predetermined method. Just do it.

  Next, the personal computer generates a calibration subband data group by performing orthogonal transformation on the calibration pitch waveform data (step S113). Then, the intensity before the change of the kth subband data in the subband data group acquired in step S101 is X (k), and the kth subband data in the calibration subband data group generated in step S113. In the subband data group so that the intensity of the k-th subband data in the subband data group acquired in step S101 becomes the above-described value Y (k) shown in Equation 3. Are changed (step S114), and the process proceeds to step S116. Note that, in the state where the calibration subband data group has not yet been created, the personal computer may treat the subband data group acquired in step S101 as having undergone the process of step S114.

  Next, when the sound quality designation data is also acquired, this personal computer further converts the intensity of each subband data in the subband data group after changing the value in step S114 to the intensity designated by the sound quality designation data. By changing, the sound quality of the sound represented by the subband data group as a whole is adjusted (step S115), and the process proceeds to step S116.

  In step S116, the personal computer converts the subband data group that has undergone the processing up to step S114 or S115, thereby converting pitch waveform data or audio data in which the intensity of each frequency component is represented by the subband data group. Restore. The conversion applied to the subband data group in step S116 is a conversion that has a substantially inverse relationship to the conversion performed on the audio data to generate the subband data group acquired in step S101. And

  Next, this personal computer changes the time length of each section of the pitch waveform data to be the time length indicated by the pitch information acquired in step S101 if the data generated in step S116 is pitch waveform data. (Step S117), and the process proceeds to Step S118. On the other hand, if the data generated in step S116 is audio data that has not undergone the process of aligning the phases of the sections, the process of step S117 is omitted, and the process immediately proceeds to step S118.

  In step S118, the personal computer demodulates the audio data obtained by the processing up to step S116 or step S117, performs D / A conversion and amplification, and reproduces the audio using the obtained analog signal.

For example, this program may be uploaded to a bulletin board (BBS) on a communication line and distributed via the communication line. Also, a carrier wave is modulated with a signal representing this program, and the obtained modulated wave is transmitted. The apparatus that receives the modulated wave may demodulate the modulated wave to restore these programs.
The above-described processing can be executed by starting this program and executing it under the control of the OS in the same manner as other application programs.

  When the OS shares a part of the processing, or when the OS constitutes a part of one component of the present invention, a program excluding the part is stored in the recording medium. May be. Also in this case, in the present invention, it is assumed that the recording medium stores a program for executing each function or step executed by the computer.

It is a block diagram which shows the structure of the sound quality adjustment apparatus which concerns on embodiment of this invention. It is a block diagram which shows the structure of a pitch extraction part. It is a flowchart which shows the process which the personal computer which performs the function of the audio | voice adjustment apparatus which concerns on embodiment of this invention performs. It is a continuation of the flowchart which shows the process which the personal computer which performs the function of the audio | voice adjustment apparatus which concerns on embodiment of this invention performs.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Subband data input part 2 Sound quality designation | designated data input part 3 Calibration data generation part 31 Calibration voice input part 32 Pitch extraction part 321 Cepstrum analysis part 322 Autocorrelation analysis part 323 Weight calculation part 324 BPF coefficient calculation part 325 Band pass filter 326 Zero cross analysis unit 327 Waveform correlation analysis unit 328 Phase adjustment unit 329 Resampling unit 33 Subband analysis unit 4 Sound quality adjustment unit 5 Audio reproduction unit 51 Subband synthesis unit 52 Audio waveform restoration unit 53 Audio output unit

Claims (6)

Audio signal acquisition means for acquiring an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of audio from the outside;
An audio signal adjustment unit that changes the intensity of an audio signal included in the audio signal group acquired by the audio signal acquisition unit;
Waveform generating means for generating a signal representing a waveform of a voice represented by the voice signal group based on the voice signal group whose intensity of the voice signal is changed,
An audio signal adjustment device characterized by the above. Further comprising designation data obtaining means for obtaining designation data for designating a mode of change in intensity of the audio signal included in the audio signal group obtained by the audio signal obtaining means from the outside;
The audio signal adjustment unit changes the intensity of the audio signal included in the audio signal group acquired by the audio signal acquisition unit in a manner specified by the specified data acquired by the specified data acquisition unit.
The audio signal adjusting apparatus according to claim 1, wherein Calibration audio signal that receives audio and generates a calibration audio signal group consisting of calibration audio signals that represent temporal changes in the intensity of the fundamental frequency component and the harmonic component of the processing target audio signal that represents the waveform of the audio. Further comprising generating means,
The audio signal adjustment means uses the value after the change of the intensity of the audio signal included in the audio signal group acquired by the audio signal acquisition means as the intensity of the audio signal and the frequency substantially the same as the audio signal. Determining based on the intensity of the calibration audio signal representing the component of, and changing the intensity of the audio signal according to the determination result,
The audio signal adjustment device according to claim 1 or 2, The calibration audio signal generation means includes:
Means for generating a signal representing the waveform of the received sound and processing the signal into a pitch waveform signal by making the time lengths of the sections corresponding to the unit pitch of the signal substantially the same;
Means for generating, as the calibration audio signal, a signal representing a temporal change in intensity of the fundamental frequency component and the harmonic component of the pitch waveform signal;
The audio signal adjusting device according to claim 3, wherein Obtain an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of the audio from the outside,
Change the intensity of the audio signal included in the acquired audio signal group,
Based on the audio signal group in which the intensity of the audio signal is changed, a signal representing the waveform of the audio represented by the audio signal group is generated.
A method of adjusting an audio signal. Computer
Audio signal acquisition means for acquiring an audio signal group consisting of audio signals representing temporal changes in the intensity of the fundamental frequency component or harmonic component of audio from the outside;
An audio signal adjustment unit that changes the intensity of an audio signal included in the audio signal group acquired by the audio signal acquisition unit;
A waveform generating means for generating a signal representing a waveform of a voice represented by the voice signal group based on the voice signal group in which the intensity of the voice signal is changed;
Program to make it function.

Images