JP2016111478A - Voice processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain noise reduction in consideration of a frequency fluctuation of a noise component due to temperature change and secular change, etc.SOLUTION: A voice processing device includes: voice input means; drive means; a memory for storing a noise profile to be generated from a noise spectrum obtained by converting the sound signal of noise related to the drive means into a frequency area; correction means for correcting the noise profile with the use of auxiliary data indicating a fluctuation due to the disturbance of noise related to the drive means; and voice processing means for reducing noise related to the drive means from a voice signal to be input by the voice input means with the use of the noise profile corrected by the correction means.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a speech processing apparatus that reduces noise contained in speech captured by a speech input means such as a microphone, and more specifically to a speech processing apparatus that reduces noise by a spectral subtract method (SS method).

  2. Description of the Related Art Conventionally, as a method for reducing noise contained in an audio signal captured by a microphone, a method using spread spectrum (SS) that selectively reduces noise frequency components is known. This is to reduce the noise by estimating the frequency component of the noise and subtracting the frequency component from the input voice signal.

  Patent Document 1 describes a noise reduction system in which a noise spectrum that can be included in an input audio signal is stored in advance as a noise profile.

JP 2011-97260 A

  Even though noise has a frequency component that is fixed to some extent, its spectrum can easily fluctuate due to various disturbances. Since the technique described in Patent Document 1 performs noise reduction using a noise profile that does not take into account frequency fluctuations of noise components due to temperature change, secular change, etc., the result of appropriately reducing noise in an actual situation is obtained. difficult. On the other hand, an unpleasant voice component may be generated.

  An object of the present invention is to provide a speech processing apparatus that can cope with noise fluctuations due to temperature change or secular change while using a noise profile.

  In order to achieve such an object, an audio processing apparatus according to the present invention is incorporated in a main apparatus including noise generating means and audio input means for taking in noise generated by the noise generating means together with other sounds. A speech processing apparatus, wherein a memory that stores a noise profile created from a noise spectrum obtained by converting the noise speech signal into a frequency domain, and fluctuation due to noise disturbance generated by the noise generating means Correction means for correcting the noise profile using auxiliary data; and voice processing means for reducing the noise contained in the voice signal input by the voice input means using the noise profile corrected by the correction means. It is characterized by comprising.

  According to the present invention, by correcting the noise profile in consideration of the fluctuation of the frequency component of noise due to disturbance, it is possible to perform noise reduction that greatly reduces the remaining noise component.

It is a schematic block diagram of an embodiment of an imaging apparatus to which the present invention is applied. It is explanatory drawing of the correction noise profile creation procedure. It is explanatory drawing of the preparation procedure of the noise profile before correction | amendment. It is a wave form diagram which shows the example of a waveform of a floor noise spectrum. It is a wave form diagram which shows the waveform example of a noise spectrum. It is a wave form diagram which shows the example of a waveform of the noise profile before correction | amendment. It is a wave form diagram which shows the relationship between the waveform example of the noise profile before correction | amendment, and threshold value (gamma). It is a wave form diagram which shows the relationship between the example of a waveform of the noise profile before correction | amendment, and threshold value (gamma) and (beta). It is explanatory drawing which shows the example of a frequency characteristic of a fluctuation | variation level. It is explanatory drawing which shows the example of a frequency characteristic of auxiliary data. It is explanatory drawing which shows another example of a frequency characteristic of auxiliary data. It is explanatory drawing of the preparation procedure of a correction noise profile. It is a wave form diagram which shows the frequency characteristic example of a correction noise profile. It is a schematic block diagram of a noise profile correction apparatus. FIG. 15 is a schematic block diagram illustrating a configuration in which the imaging apparatus illustrated in FIG. 1 is combined with the noise profile correction apparatus illustrated in FIG. 14. FIG. 16 is an explanatory diagram of a procedure for creating a correction noise profile in the configuration shown in FIG. 15.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a schematic block diagram of an imaging apparatus as a main apparatus incorporating an embodiment of a sound processing apparatus according to the present invention.

  In the imaging apparatus 100 shown in FIG. 1, the imaging unit 101 converts an optical image of a subject captured by a photographic lens into an image signal by an imaging element such as a CCD sensor or a CMOS sensor, performs analog-digital conversion, and performs an image processing unit. 102. The imaging unit 101 includes a motor and gear for driving the imaging lens, and rotates the motor and gear according to a control signal from the control unit 114, thereby performing zoom-in / zoom-out operations of the imaging lens. The motor gear and the like are driving means for driving the imaging lens.

  The image processing unit 102 performs image quality adjustment processing for adjusting white balance, color, brightness, and the like on the digital image signal input from the imaging unit 101 based on setting values. The image processing unit 102 transmits the processed image data to the memory 105, the video output unit 110, the display control unit 111, and the control unit 114 according to the situation.

  The audio input unit 103 takes in audio around the imaging device 100 via at least one or more built-in or external microphones, converts the audio into a digital signal, and supplies the digital signal to the audio processing unit 104.

  The sound processing unit 104 performs sound processing such as sound level optimization processing and specific frequency reduction processing on the sound data input from the sound input unit 103, and supplies the processing results to the memory 105 and the sound output unit 109. . The voice processing unit 104 incorporates noise reduction means. The voice processing unit 104 also uses the voice data from the voice input unit 103 to generate data such as a noise profile used for reducing noise by the SS (spread spectrum) method, and supplies the data to the memory 105. The speech processing unit 104 includes a fast Fourier transform (FFT) circuit that performs fast Fourier transform on speech data and an inverse fast Fourier transform (IFFT) circuit that performs inverse fast Fourier transform.

  The memory 105 stores data such as an image signal from the image processing unit 102, an audio signal from the audio processing unit 104, and a noise profile.

  The encoding processing unit 106 can encode and decode image data and audio data by a predetermined image encoding method and audio encoding method, respectively. The encoding processing unit 106 reads out image data and audio data temporarily stored in the memory 105, performs image encoding and audio encoding, respectively, and generates compressed image data and compressed audio data, and recording control Supplied to the unit 107.

  The recording control unit 107 records the compressed image data and compressed audio data generated by the encoding processing unit 106 and the control data related to shooting on the recording medium 108. The recording control unit 107 can also read (reproduce) compressed image data, compressed audio data, various data, and programs recorded on the recording medium 108. The recording control unit 107 supplies the read compressed image data and compressed audio data to the encoding processing unit 106. The recording medium 108 includes a general-purpose recording medium capable of recording various data, such as a magnetic disk, an optical disk, or a semiconductor memory, and may be a combination of a plurality of different or similar media.

  The encoding processing unit 106 temporarily stores the compressed image data and the compressed audio data from the recording control unit 107 in the memory 105 and decodes them according to a predetermined procedure. The encoding processing unit 106 supplies the decoded audio data to the audio output unit 109, and supplies the decoded image data to the video output unit 110 and the display control unit 111.

  The audio output unit 109 includes, for example, an audio output terminal, and outputs an audio signal to an audio output device such as an earphone or a speaker. The audio output unit 109 may be a speaker built in the imaging apparatus 100. The video output unit 110 includes a video output terminal, for example, and outputs an image signal to an external display or the like. The audio output unit 109 and the video output unit 110 may be configured as one integrated terminal, for example, a High-Definition Multimedia Interface (HDMI) (registered trademark) terminal.

  The display control unit 111 displays the image signal from the encoding processing unit 106 and the image of the image signal from the image processing unit 102 on the display unit 112, and displays an operation screen (menu screen) for operating the imaging apparatus 100. It is displayed on the display unit 112. The display unit 112 may be any display device such as a liquid crystal display, an organic EL display, or electronic paper.

  The operation unit 113 transmits an instruction signal having contents corresponding to the user operation to the control unit 114. The control unit 114 controls each block by transmitting a control signal to each block of the imaging apparatus 100 based on the instruction signal transmitted from the operation unit 113. The operation unit 113 includes, for example, a power button, a recording start button, a menu display button, a determination button, a cursor key, a pointing device for designating an arbitrary point on the display unit 112, a touch panel, and the like.

  The control unit 114 includes, for example, a CPU (MPU) and a memory (DRAM, SRAM) for executing various processes (programs), and controls operations related to shooting of the imaging apparatus 100. For example, upon receiving a zoom shooting instruction signal from the operation unit 113, the control unit 114 transmits a control signal indicating a zoom operation to the imaging unit 101.

  The communication unit 115 communicates with an external device wirelessly or by wire and transmits / receives data such as audio data and image data to / from the external device. The communication unit 115 also transmits / receives a control signal related to shooting, such as a shooting start / end command, and other information. The communication unit 115 includes, for example, one or more of an infrared communication module, a Bluetooth (registered trademark) communication module, a wireless LAN communication module, a wired LAN communication module, a USB communication module, and a Thunderbolt (registered trademark) communication module. The communication unit 115 also serves as an interface for connecting to a device such as a remote operation device (remote controller) or a personal computer (PC).

  The bus 116 is a medium that transmits various data, control signals, and the like between the blocks of the imaging apparatus 100.

  The moving image / sound recording operation of this embodiment will be described. When the user selects the moving image shooting mode with the operation unit 113 and presses the REC button, the control unit 114 transmits an operation recording start control signal to the related block in response to the operation signal from the operation unit 113. In response to the control signal for starting moving image shooting from the control unit 114, the imaging unit 101 starts moving image shooting. The audio input unit 103 receives a control signal for starting moving image shooting from the control unit 114 and starts to capture surrounding audio. The moving image shot by the imaging unit 101 is subjected to image signal processing such as color temperature and white balance adjustment by the image processing unit 102 and temporarily stored in the memory 105 as moving image data.

  When the user performs a zoom operation with the operation unit 113 during moving image recording, the control unit 114 supplies a control signal for starting the zoom operation to the imaging unit 101. The imaging unit 101 receives a zoom start control signal from the control unit 114, rotates a motor / gear to move the imaging lens, and generates noise due to the motor / gear to move the imaging lens. Noise generated by motors, gears, etc. for moving the imaging lens is called “zoom noise”. When zoom noise occurs, the audio signal captured by the audio input unit 103 includes moving image audio and zoom noise associated with the captured moving image. The audio input unit 103 inputs an audio signal in which moving image audio and zoom noise are mixed into the audio processing unit 104.

  The audio processing unit 104 performs audio signal processing such as level adjustment on the audio signal input from the audio input unit 103, and temporarily stores the processed audio data in the memory 105. Upon receiving a zoom operation control signal from the control unit 114, the audio processing unit 104 starts noise reduction processing for reducing zoom noise in addition to audio signal processing during moving image shooting. That is, the control unit 114 activates the noise reduction unit of the audio processing unit 104 while zoom noise is generated. While receiving a control signal for zoom operation during moving image shooting from the control unit 114, the sound processing unit 104 executes noise reduction processing by the SS method using the correction noise profile.

  The encoding processing unit 106 compresses and encodes moving image data and audio data temporarily stored in the memory 105. The recording control unit 107 records the compressed moving image data and the compressed audio data from the encoding processing unit 106 on the recording medium 108 as an AV (Audio Visual) file.

  The noise reduction processing in the voice processing unit 104 will be described in detail. The audio processing unit 104 performs fast Fourier transform (FFT) on the audio data input from the audio input unit 103 (audio signal in which moving image audio and zoom noise are mixed) to obtain an input audio spectrum. In accordance with the zoom operation control signal from the control unit 114, the sound processing unit 104 selects and reads the optimum correction noise profile stored in the memory 105. The sound processing unit 104 subtracts the corrected noise profile read from the memory 105 from the input sound spectrum, and performs inverse fast Fourier transform on the subtraction result. By this process, the noise component contained in the input voice signal can be reduced. As described above, the audio signal with reduced noise components is processed by the encoding processing unit 106 and the recording control unit 107 and recorded on the recording medium 108 as described above.

  As shown in FIG. 2, the audio processing unit 104 subtracts the floor noise spectrum from the noise spectrum to generate a noise profile before correction. Then, the sound processing unit 104 corrects the pre-correction noise profile using auxiliary data to generate a corrected noise profile. The floor noise spectrum is a spectrum of background noise generated by the imaging apparatus 100 itself, and the noise noise is a state in which noise to be reduced (here, zoom noise generated by the imaging unit 101) is added to the background noise. The spectrum of noise is shown. By subtracting the floor noise spectrum from the noise spectrum, a spectrum of only noise to be reduced is obtained, and this becomes a noise profile before correction. Auxiliary data is data that reflects changes in frequency characteristics due to disturbance of noise to be reduced. The corrected noise profile obtained by correcting the pre-correction noise profile with this auxiliary data is a noise profile suitable for noise reduction at present.

  With reference to FIG. 3, a procedure for creating a noise profile before correction will be described. The imaging device 100 is installed in an anechoic soundproof box. An anechoic soundproof box is a sealable box or room in which sound inside does not reverberate and outside sound does not enter inside in a sealed state. The imaging device 100 is connected to a remote controller outside the anechoic soundproof box by wireless or wired connection, and is installed in a state where it can be operated from the outside of the anechoic soundproof box. All operations of the imaging apparatus 100 installed in the anechoic soundproof box are performed by a remote controller connected to the communication unit 115.

  An instruction signal (for example, a recording start instruction signal) is input from the remote controller to the control unit 114 via the communication unit 115 of the imaging apparatus 100. In response to this instruction signal, the control unit 114 causes the audio input unit 103 to start capturing audio. The audio input unit 103 captures audio for one second, for example, in a state where the imaging unit 101 is not performing a zoom operation. The audio processing unit 104 performs fast Fourier transform on the input audio signal for one second. Thereby, a spectrum of noise (background noise) generated by the image capturing apparatus 100 itself is obtained. The average value at each frequency of the spectrum obtained for 1 second in this way is defined as a floor noise spectrum (FIG. 4). The audio processing unit 104 writes the obtained floor noise spectrum in the memory 105 and stores it.

  Next, in the sealed anechoic soundproof box, the control unit 114 rotates the motor / gear of the imaging unit 101 and the like during zooming while the voice input unit 103 inputs voice. Two zoom operations are performed from the wide-angle end to the telephoto end and from the telephoto end to the wide-angle end. For example, it is assumed that zoom operations from the wide-angle end to the telephoto end and from the telephoto end to the wide-angle end take exactly 4 seconds, respectively. The audio processing unit 104 performs a fast Fourier transform on a 4-second audio signal obtained by the audio input unit 103 during the zoom operation from the wide-angle end to the telephoto end (except for the startup sound that appears only when the zoom operation starts). Thereby, a spectrum of noise generated by the zoom operation from the wide angle end to the telephoto end can be obtained. As shown in FIG. 5, the voice processing unit 104 sets the peak value at each frequency of the spectrum as a noise spectrum, and temporarily stores the noise spectrum in the memory 105. Similarly, the sound processing unit 104 temporarily stores in the memory 105 a noise spectrum of noise generated by a zoom operation from the telephoto end to the wide-angle end.

  The auxiliary data is created from the variation range and variation range (variation level) of the zoom noise due to disturbance at each frequency in the frequency domain, and is used for correcting the noise profile before correction. The disturbance is, for example, due to a change in temperature or humidity, an aging deterioration of the imaging device, or the like. The noise spectrum includes a spectrum that varies in the frequency axis direction due to these disturbances and a spectrum that does not vary. The noise spectrum that fluctuates in the frequency axis direction causes unerased noise components in the noise reduction by the SS method.

  The audio processing unit 104 (or the control unit 114) calculates auxiliary data and stores it in the memory 105 as described below. The disturbance as described above is applied to the imaging apparatus 100, and the sound processing unit 104 is caused to calculate the noise profile before correction as described above. FIG. 6 shows an example of frequency characteristics of the noise profile before correction for the floor level spectrum shown in FIG. 4 and the noise spectrum shown in FIG.

  For example, a pre-correction noise profile for each temperature and humidity is obtained in an anechoic soundproof box that can change temperature and humidity. The control unit 114 compares the obtained plurality of pre-correction noise profiles, and detects a spectrum of a threshold value γ or more that varies in the frequency axis direction due to changes in temperature and humidity. Here, as an example, the threshold γ is an average value of the sum of the noise profile before correction in the frequency axis direction, but this is experimentally determined, and the threshold γ is not limited to this.

周波数ｘ（Ｈｚ）に対しての補正パラメータαを得る。補正パラメータαを周波数順に並べたものが補助データとなる。得られた補助データの周波数分布を図１０に示す。つまり、補助データは、ある周波数ｘ（Ｈｚ）に対してただ一つの補正パラメータαが求まるデータのことである。 The control unit 114 creates auxiliary data from the fluctuation range and fluctuation level in which the spectrum equal to or greater than the threshold γ fluctuates in the frequency axis direction, and stores the created auxiliary data in the memory 105 so that the voice processing unit 104 can access the auxiliary data. . For example, in the noise profile before correction, as shown in FIG. 7, it is assumed that there is a spectrum having a threshold value γ or more standing around 3000 Hz on the frequency axis. It is assumed that this spectrum fluctuates between 2900 Hz and 3100 Hz when the temperature or humidity is changed. As shown in FIG. 9, the fluctuation range at this time is 2900 (Hz) for the minimum value (min_freq) and 3100 (Hz) for the maximum value (max_freq). The fluctuation level (lev_freq = max_freq−min_freq) is 200 (Hz). By substituting these values into the following equation:

A correction parameter α for the frequency x (Hz) is obtained. The auxiliary data is obtained by arranging the correction parameters α in order of frequency. The frequency distribution of the auxiliary data obtained is shown in FIG. That is, auxiliary data is data for which only one correction parameter α is obtained for a certain frequency x (Hz).

  For disturbances other than temperature and humidity, the control unit 114 also compares the pre-correction noise profiles obtained under actual disturbance conditions or conditions that can obtain results equivalent to the actual disturbance conditions. Gain levels and create auxiliary data.

  The correction parameter α may be obtained by another formula. Depending on the noise reduction conditions, the correction parameter α may be a frequency distribution as shown in FIG. The present embodiment is characterized in that auxiliary data is created from a fluctuation range and a fluctuation level, and the pre-correction noise profile is corrected using the auxiliary data. The correction parameters illustrated in FIGS. 10 and 11 are used. It is not limited to α.

  As shown in FIG. 12, the sound processing unit 104 generates a corrected noise profile by correcting the pre-correction noise filter corresponding to the operating state with auxiliary data corresponding to the current disturbance. The pre-correction noise profile necessary for creating the correction noise profile is not necessarily the same as the pre-correction noise profile used when creating the auxiliary data.

  In the frequency band where the correction parameter α of the auxiliary data is α = 0 (Hz), the pre-correction noise profile is not corrected. Even when the value of the noise profile before correction is equal to or less than the threshold value γ, the noise profile before correction is not corrected. In either case, the value of the pre-correction noise profile becomes the value of the correction noise profile as it is. That is, only when the correction parameter α is a frequency band where α ≠ 0 and the pre-correction noise profile takes a value greater than or equal to the threshold γ, the value obtained by correcting the pre-correction noise profile using auxiliary data is the correction noise. The profile value.

For example, as shown in FIG. 8, consider a pre-correction noise profile in which a spectrum stands with a peak at 2965 Hz within a range of min_freq to max_freq. At this time, all spectra existing in the range of min_freq to max_freq are defined as a spectrum group A, and the peak value of the spectrum group A is defined as β (dB). In this case, the correction parameter α is determined by the above-described α calculation formula.
α = 200- (2965- (3100 + 2900) / 2) ² × 2/200 = 187.75
Is calculated.

  The spectrum group A is moved in the frequency axis direction within a range of width α = 187.75 (Hz) (2871.125 (Hz) to 3058.875 (Hz)) centering on 2965 Hz which is a peak frequency. A peak obtained when the spectrum group A moves is a corrected noise profile. FIG. 13 shows the frequency characteristics of the correction noise profile obtained in this way. A part of the moved spectrum group A goes out of the range of min_freq to max_freq. At this time, outside the range of min_freq to max_freq, the value obtained by taking the peak when the spectrum group A moves is compared with the value of the noise profile before correction, and the larger value is set as the correction noise profile value.

  By performing noise reduction by the SS method using the correction noise profile obtained in this way, in the noise reduction, for example, it is possible to greatly reduce the remaining noise components due to changes in temperature and humidity, aging, and the like.

  FIG. 14 is a block diagram illustrating a schematic configuration of a noise profile correction apparatus that corrects the noise profile of the imaging apparatus 100. A noise profile correction apparatus 1400 shown in FIG. 14 includes an arithmetic device 1410, an audio device 1417 connected to the arithmetic device 1410, a video display device 1418, and an operation device 1419.

  The computing device 1410 is composed of, for example, a personal computer or a notebook computer. The central arithmetic control unit 1411 is, for example, a CPU inside the personal computer, executes a predetermined program developed on the memory 1412, and controls each block of the arithmetic device 1410 in accordance with an instruction signal from the operation device 1419. The memory 1412 temporarily stores data, programs, and the like used by the arithmetic device 1410.

  A communication unit 1413 communicates with an external device wirelessly or by wire and transmits / receives data such as audio data and image data to / from the external device. The communication unit 1413 also transmits / receives an instruction signal to the arithmetic device 1410 and other information to / from an external device. The communication unit 1413 includes, for example, one or more of an infrared communication module, a Bluetooth (registered trademark) communication module, a wireless LAN communication module, a wired LAN communication module, a USB communication module, and a Thunderbolt (registered trademark) communication module. The communication unit 115 is an interface for connecting to an external digital device such as a digital camera or a device such as a mouse or a keyboard.

  The recording medium 1414 includes a general-purpose recording medium capable of recording various data, such as a magnetic disk, an optical disk, or a semiconductor memory, and may be a combination of a plurality of different or similar media.

  The audio output unit 1415 includes, for example, an audio output terminal, and outputs an audio signal to an audio output device such as an earphone or a speaker. The audio output unit 1415 may be a speaker built in the arithmetic device 1410. The video output unit 1416 includes a video output terminal, for example, and outputs an image signal to an external display or the like. The audio output unit 1415 and the video output unit 1416 may be configured as a single integrated terminal, for example, a High-Definition Multimedia Interface (HDMI) (registered trademark) terminal.

  The acoustic device 1417 is an acoustic device connected to the arithmetic device 1410 by wire or wireless, and is, for example, an earphone or a speaker. The audio equipment 1417 may be a USB audio device and a plurality of audio equipment such as a headphone amplifier and headphones attached thereto.

  The video display device 1418 is a video display device that is wired or wirelessly connected to the arithmetic device 1410 and may be any display device such as a liquid crystal display, an organic EL display, or electronic paper.

  The operation device 1419 is an operation device connected to the arithmetic device 1410 by wire or wireless, and may be any operation device such as a mouse, a keyboard, or a touch panel. Further, the operation device 1419 may be a plurality of devices.

  The operation of creating a corrected noise profile by the noise profile correction apparatus 1400 will be described. As illustrated in FIG. 15, the noise profile correction device 1400 is connected to the imaging device 100 by connecting the communication unit 1413 of the noise profile correction device 1400 and the communication unit 115 of the imaging device 100. The noise profile correction apparatus 1400 stores the auxiliary data described above in the memory 1412.

  As described above, the imaging apparatus 100 is installed in an anechoic soundproof box, and a pre-correction noise profile is created and stored in the memory 105 with the anechoic soundproof box sealed. The arithmetic device 1410 transfers the pre-correction noise profile stored in the memory 105 to the memory 1412 via the communication unit 115 and the communication unit 1413.

  The central processing control unit 1411 reads auxiliary data and a pre-correction noise profile from the memory 1412, and creates a correction noise profile by the same calculation procedure shown in FIG. 16 as described above. The central processing control unit 1411 temporarily stores the obtained corrected noise profile in the memory 202 and transfers it to the memory 105 of the imaging apparatus 100 via the communication unit 1413 and the communication unit 115.

  In this way, the noise profile correction apparatus 1400 can correct the pre-correction noise profile stored in the memory 105 of the imaging apparatus 100 with the auxiliary data and write it back to the memory 105 of the imaging apparatus 100. The audio processing unit 104 of the imaging apparatus 100 performs noise reduction by the SS method using the corrected noise profile transmitted from the noise profile correcting apparatus 1400. This makes it possible to reduce noise by greatly reducing unremoved noise components due to, for example, changes in temperature, humidity, and aging.

(Other examples)
The sound processing apparatus according to the present invention is not limited to the imaging apparatus 100 described in the first embodiment. For example, the audio processing apparatus according to the present invention can be realized by a system including a plurality of apparatuses.

  The various processes and functions described in the first embodiment can also be realized by a computer program. In this case, the computer program according to the present invention can be executed by a computer (including a CPU and the like), and implements various functions described in this embodiment.

  It goes without saying that the computer program according to the present invention may realize various processes and functions described in this embodiment by using an OS (Operating System) running on the computer.

  The computer program according to the present invention is read from a computer-readable recording medium and executed by the computer. As the computer-readable recording medium, a hard disk device, an optical disk, a CD-ROM, a CD-R, a memory card, a ROM, or the like can be used. The computer program according to the present invention may be provided from an external device to a computer via a communication interface and executed by the computer.

Claims (4)

Voice input means;
Driving means;
A memory for storing a noise profile created from a noise spectrum obtained by converting an audio signal of noise related to the driving means into a frequency domain;
Correction means for correcting the noise profile using auxiliary data indicating fluctuation due to noise disturbance related to the driving means;
An audio processing apparatus comprising: an audio processing unit that reduces noise related to the driving unit from an audio signal input by the audio input unit using the noise profile corrected by the correction unit. The sound processing device is an imaging device;
The audio processing apparatus according to claim 1, wherein the driving unit is a unit that drives a zoom.   The speech processing apparatus according to claim 1, wherein the auxiliary data includes data indicating a fluctuation range and a fluctuation range of the noise in a frequency axis direction.   The voice processing apparatus according to claim 1, further comprising a control unit that activates the voice processing unit while the driving unit generates the noise.

Images