JP6102053B2 - Sound processing apparatus and sound processing method - Google Patents

Description

  The present invention relates to a technique for processing an acoustic signal.

  Conventionally, a technique for suppressing howling that occurs in an acoustic system including a sound collecting device and a sound emitting device has been proposed. For example, Patent Literature 1 discloses a technique for suppressing howling using an adaptive filter that simulates a transmission path from a sound emitting device to a sound collecting device.

JP 2006-203553 A

  However, in a technique for suppressing howling using an adaptive filter as in the technique of Patent Document 1, a considerable amount of time is required from howling to convergence of a coefficient group of the adaptive filter. Therefore, for example, when a significant characteristic variation occurs in the transmission path in a short time, the coefficient group of the adaptive filter may not reach convergence, and as a result, howling may not be sufficiently suppressed. In view of the above circumstances, an object of the present invention is to quickly suppress howling.

  The acoustic processing apparatus of the present invention uses reverberation analysis means (for example, reverberation analysis unit 24A) that estimates a frequency in which a reverberation component of an observation signal (for example, observation signal x (t)) exists as a reverberation frequency (for example, reverberation frequency FR (m)). , 24B) and band suppression means (for example, band suppression unit 22) for suppressing the band component including the reverberation frequency in the observation signal. With the above configuration, since the reverberation component that causes howling is suppressed in the observation signal, howling caused by the reverberation component can be prevented in advance.

  In the first aspect of the present invention, the reverberation analysis means estimates a peak frequency of a reverberation component spectrum (for example, a reverberation spectrum YR (k, m)) included in the observation signal as a reverberation frequency. In the above aspect, since the peak frequency of the reverberation component spectrum is estimated as the reverberation frequency, there is an advantage in that howling can be effectively prevented by suppressing components that tend to cause howling among the reverberation components. Specifically, the reverberation analysis means calculates the reverberation dominance (for example, reverberation dominance Gr (k, m)) corresponding to the ratio of the reverberation component of the observation signal for each frequency (for example, the analysis processing unit 34). ) And reverberation extraction means (for example, the reverberation extraction unit 36) for extracting the reverberation component by applying the reverberation dominance of each frequency to the observation signal, and the reverberation of the peak frequency of the reverberation component spectrum extracted by the reverberation extraction means Frequency specifying means (for example, frequency specifying unit 38) that specifies the frequency.

  In the second aspect of the present invention, the analysis processing means for calculating the reverberation dominance corresponding to the ratio of the reverberation component of the observation signal for each frequency, and the peak frequency of the reverberation dominance on the frequency axis are specified as the reverberation frequency. Frequency specifying means. In the above aspect, since the peak frequency of the reverberation dominance is specified as the reverberation frequency, there is an advantage that the process of extracting the reverberation component from the observation signal is unnecessary.

  In the first aspect and the second aspect, the analysis processing means is lower than the first index value (for example, the first index value Q1 (k, m)) that follows the time change of the observation signal and the first index value. Index value calculation means (for example, index value calculation units 42A and 42B) for calculating a second index value (for example, the second index value Q2 (k, m)) that follows the time change of the observation signal with tracking capability; And a dominance degree calculating means (for example, dominance degree calculating unit 44) for calculating a reverberation dominance degree according to the difference between the index value and the second index value. In the above aspect, since the reverberation dominance is calculated according to the difference between the first index value and the second index value following the time change of the observation signal, for example, the prediction filter for estimating the reverberation component included in the observation signal By using a coefficient probability model, the reverberation component of the observation signal can be estimated with a simpler process compared to a configuration that estimates the prediction filter coefficient of the reverberation component (for example, a configuration disclosed in Japanese Patent Laid-Open No. 2009-212599). There is an advantage. However, a known technique (including the configuration disclosed in Japanese Patent Laid-Open No. 2009-212599) can be arbitrarily employed for estimating the reverberation component in the present invention.

  In a specific aspect, the index value calculation means is a first smoothing means (for example, a first smoothing means for calculating the first index value by smoothing the time series of the signal strength of the observation signal (the amplitude of the observation signal or its power). 1 smoothing unit 51) and the time series of the signal strength of the observation signal by the time constant (eg, time constant τ2) exceeding the time constant (eg, time constant τ1) of smoothing by the first smoothing means 2nd smoothing means (for example, 2nd smoothing part 52) which calculates an index value. In another aspect, the index value calculating means smooths the time series of the signal strength of the observation signal so that the time change of the second index value has a relationship that the time change of the first index value is delayed. An index value and a second index value are generated.

  In a preferred aspect of the present invention, the dominance calculation means includes a ratio calculation means (for example, a ratio calculation unit 62) for calculating a ratio of the first index value to the second index value, and the predetermined degree when the ratio exceeds a predetermined value. The first processing means (for example, the first processing unit 64) that calculates the initial sound dominance (for example, the initial sound dominance Ge (k, m)) set to the ratio when the ratio is lower than the predetermined value. And second processing means (for example, the second processing unit 66) for calculating the reverberation dominance by subtracting the initial sound dominance from a predetermined value. The above aspect has an advantage that the reverberation dominance can be calculated by a simple calculation including the calculation of the ratio of the first index value to the second index value and the calculation of subtracting the initial sound dominance from the predetermined value.

  The acoustic processing device according to a preferred aspect of the present invention includes reverberation suppression means (for example, a reverberation suppression unit 72) that suppresses a reverberation component of an observation signal, and the band suppression means is before or after processing by the reverberation suppression means. The band component including the reverberation frequency is suppressed among the observation signals. In the above aspect, since the reverberation suppression unit suppresses the reverberation component of the observation signal, it is possible to suppress the increase in howling over time for the band components other than the reverberation frequency. In the configuration including the first processing means for calculating the initial sound dominance, the reverberation suppressing means suppresses the reverberation component by applying the initial sound dominance of each frequency calculated by the first processing means to the observation signal. A configuration is preferred. In the above configuration, since the initial sound dominance is diverted to the calculation of the reverberation dominance by the second processing means and the suppression of the reverberation component by the dereverberation suppression means, the processing is performed in comparison with the configuration in which each process is executed independently. Has the advantage of being simplified.

  The acoustic processing apparatus according to a preferred aspect of the present invention includes delay means (for example, a delay unit 74) for delaying the observation signal, and the reverberation analysis means uses the observation signal before processing by the delay means as a processing target, and performs band suppression. The means uses the observation signal processed by the delay means as a processing target. In the above configuration, even when time is required for the calculation of the reverberation frequency by the reverberation analysis means, the analysis processing means identifies the time point of the observation signal when the band suppression means processes the specific time point of the observation signal. It is possible to apply the reverberation frequency to the band suppression means. Therefore, there is an advantage that howling due to the reverberation component can be effectively prevented.

  The present invention can also be specified as a device (howling frequency specifying device) that specifies a reverberation frequency of a band component to be suppressed in order to prevent howling due to a reverberant component. The howling frequency specifying device of the present invention is a reverberation analysis unit (for example, a reverberation analysis unit) that estimates a frequency in which a reverberation component of an observation signal (for example, an observation signal x (t)) exists as a reverberation frequency (for example, a reverberation frequency FR (m)). 24A, 24B). The sound processing apparatus of the present invention can be configured by adding band suppression means for suppressing band components including the reverberation frequency in the observation signal to the howling frequency specifying apparatus.

  The acoustic processing device according to each of the above aspects is realized by hardware (electronic circuit) such as a DSP (Digital Signal Processor) dedicated to processing of an acoustic signal, or a general-purpose calculation such as a CPU (Central Processing Unit). This is also realized by cooperation between the processing device and the program. The program according to the present invention causes a computer to execute a reverberation analysis process for estimating a frequency at which a reverberation component of an observation signal exists as a reverberation frequency, and a band suppression process for suppressing a band component including the reverberation frequency in the observation signal. According to the above program, the same operation and effect as the sound processing apparatus according to the present invention are realized. Note that the program of the present invention is provided in a form stored in a computer-readable recording medium and installed in the computer, or is provided in a form distributed via a communication network and installed in the computer.

1 is a block diagram of a loudspeaker according to a first embodiment of the present invention. It is a block diagram of a reverberation analysis part. It is a block diagram of an analysis processing part. It is explanatory drawing of the relationship between a 1st index value, a 2nd index value, and an initial sound dominance degree. It is a block diagram of the reverberation analysis part in 2nd Embodiment. It is a block diagram of the sound processing apparatus in 3rd Embodiment. It is a block diagram of the sound processing apparatus in 4th Embodiment. It is a block diagram of the analysis processing part in a 5th embodiment. It is explanatory drawing of the relationship between the 1st parameter | index value in 2nd Embodiment, a 2nd parameter | index value, and initial stage sound dominance.

<First Embodiment>
FIG. 1 is a block diagram of a loudspeaker 10 using a sound processing apparatus 100 according to the first embodiment of the present invention. The loudspeaker 10 is a device that radiates by adjusting the volume of surrounding sound (speech and music), and includes a sound collection device 12, a sound emission device 14, and a sound processing device 100.

  The sound collection device 12 (for example, a microphone) generates an observation signal x (t) corresponding to the surrounding sound and supplies it to the sound processing device 100. The observation signal x (t) is a reverberation (initial reflection sound and initial reflection sound) that reaches the sound collecting device 12 after reflection in the acoustic space, with respect to the direct sound that directly arrives from the sound emitting device 14 to the sound collecting device 12. It is an acoustic signal in a time domain indicating a sound waveform to which (rear reverberation sound) is added (t: time).

  The acoustic processing device 100 generates an acoustic signal z (t) from the observation signal x (t) and outputs it to the sound emitting device 14. The sound emitting device 14 (for example, a speaker) emits a sound wave corresponding to the acoustic signal z (t) supplied from the acoustic processing device 100. An A / D converter that converts the observation signal x (t) generated by the sound collection device 12 from analog to digital, or a D that converts the acoustic signal z (t) generated by the sound processing apparatus 100 from digital to analog. The illustration of the / A converter is omitted for convenience.

  In a looped acoustic system in which a part of the sound wave radiated from the sound emitting device 14 (return sound) reaches the sound collecting device 12, howling may occur when the gain of the entire acoustic system exceeds 1. The acoustic processing device 100 according to the first embodiment functions as a howling suppression device that generates the acoustic signal z (t) by executing processing for preventing howling on the observation signal x (t). The sound processing device 100 is realized by, for example, a general-purpose arithmetic processing device (CPU) that executes a program or an electronic circuit (DSP) dedicated to processing of the observation signal x (t).

  Since howling is affected by the acoustic characteristics (reflection characteristics and delay characteristics in the space) of the transmission path until the radiated sound from the sound emitting device 14 reaches the sound collecting device 12, the observation signal x (t) Reverberation components (especially rear reverberation) can be an important factor in promoting the occurrence and increase of howling. Therefore, there is a tendency that howling can be effectively prevented and increased by suppressing the reverberation component of the observation signal x (t). In consideration of the above tendency, in the first embodiment, howling is prevented by suppressing a band component having a high reverberation component intensity in the observation signal x (t).

  As shown in FIG. 1, the sound processing apparatus 100 includes a band suppression unit 22, a reverberation analysis unit 24 </ b> A, and an amplification unit 26. The reverberation analysis unit 24A uses the frequency (hereinafter referred to as “reverberation frequency”) FR (m) in which the reverberation component (rear reverberation sound) of the observation signal x (t) generated by the sound collection device 12 is present as a unit period on the time axis. Estimate every frame. The symbol m means any one unit period on the time axis (a specific time point on the time axis). The band suppression unit 22 in FIG. 1 suppresses the band component including the reverberation frequency FR (m) estimated by the reverberation analysis unit 24A in the observation signal x (t) supplied from the sound collecting device 12 to thereby generate the acoustic signal z. Generate (t). Specifically, a notch filter whose frequency characteristic (stop band) is variably set so that a narrow band component centered on the reverberation frequency FR (m) is attenuated is preferably used as the band suppression unit 22. The amplification unit 26 amplifies the acoustic signal z (t) processed by the band suppression unit 22. The gain of the amplifying unit 26 is variably set according to an instruction from the user, for example. The acoustic signal z (t) processed by the amplifying unit 26 is supplied to the sound emitting device 14 and is emitted as a sound wave.

  FIG. 2 is a block diagram of the reverberation analyzer 24A. As shown in FIG. 2, the reverberation analysis unit 24 </ b> A includes a frequency analysis unit 32, an analysis processing unit 34, a reverberation extraction unit 36, and a frequency specifying unit 38. The frequency analysis unit 32 sequentially generates a spectrum (complex spectrum) X (k, m) of the observation signal x (t) for each unit period on the time axis. The symbol k is a variable that designates any one frequency (band) on the frequency axis. For generation of the spectrum X (k, m), a known frequency analysis such as a short-time Fourier transform can be arbitrarily employed. Note that a filter bank in which a plurality of bandpass filters having different passbands are arranged can be used as the frequency analysis unit 32.

  The analysis processing unit 34 calculates a reverberation dominance Gr (k, m) corresponding to the spectrum X (k, m) of the observation signal x (t) for each frequency for each unit period. The reverberation dominance Gr (k, m) of the first embodiment is a variable serving as an index of the ratio of the reverberation component (rear reverberation sound) in the observation signal x (t). Schematically, in the spectrum X (k, m) of the observed signal x (t), the higher the reverberation component intensity is, the higher the reverberation component intensity is relative to the intensity of the initial sound component (direct sound and initial reflected sound) other than the reverberation component. Gr (k, m) is set to a large value. That is, the reverberation dominance Gr (k, m) can also be expressed as the dominance of the reverberation component in the spectrum X (k, m) of the observation signal x (t).

  The reverberation extraction unit 36 applies the reverberation dominance Gr (k, m) calculated by the analysis processing unit 34 to the observation signal x (t), thereby causing the spectrum of the reverberation component in the observation signal x (t) (hereinafter referred to as “reverberation”). YR (k, m) is generated. That is, the reverberation extraction unit 36 extracts a reverberation component included in the observation signal x (t). Generation of the reverberation spectrum YR (k, m) by the reverberation extraction unit 36 is sequentially executed for each frequency for each unit period. Specifically, the reverberation extraction unit 36 reverberation dominance Gr corresponding to the spectrum X (k, m) and the unit period and frequency common to the spectrum X (k, m) of the observed signal x (t). The reverberation spectrum YR (k, m) is calculated by multiplying (k, m) (YR (k, m) = Gr (k, m) X (k, m)). That is, the reverberation dominance Gr (k, m) corresponds to a gain for the spectrum X (k, m) of the observation signal x (t), and is used to extract (emphasize) the reverberation component of the observation signal x (t). Used as a coefficient (adjustment value for reverberation enhancement).

  The frequency specifying unit 38 sequentially specifies a frequency at which a peak exists in the reverberation spectrum YR (k, m) generated by the reverberation extracting unit 36 as a reverberation frequency FR (m) for each unit period. For example, the frequency specifying unit 38 specifies one frequency at which the intensity of the reverberation spectrum YR (k, m) is maximum as the reverberation frequency FR (m). In order to suppress instantaneous fluctuations of the reverberation frequency FR (m) caused by noise or the like, the reverberation spectrum YR (k, m) is smoothed in the time axis direction, and the reverberation frequency FR ( A configuration that specifies m) is also suitable. As described above, the band component including the reverberation frequency FR (m) specified by the frequency specifying unit 38 in the observed signal x (t) is suppressed by the band suppressing unit 22 in FIG.

  As described above, in the first embodiment, reverberation components that cause howling in the observation signal x (t) are suppressed. In other words, in contrast to the technique of Patent Document 1 that suppresses howling after generation, in the first embodiment, the reverberation component that causes the suppression is suppressed before howling actually occurs, resulting in howling in advance. Is prevented. Therefore, according to the first embodiment, howling can be quickly suppressed as compared with the technique of Patent Document 1.

  The band suppression unit 22 uses the peak frequency of the spectrum X (k, m) of the observation signal x (t) including both the initial sound component (direct sound and initial reflected sound) and the reverberation component as the reverberation frequency FR (m). A configuration (hereinafter referred to as “proportional”) applied to the above can also be assumed. However, since the frequency of the peak of the spectrum differs between the initial sound component and the reverberation component, the peak frequency of the initial sound component is specified as the reverberation frequency FR (m) in comparison, and as a result, the observation signal x ( There is a possibility that the initial sound component that should be originally maintained in t) is suppressed by the band suppressing unit 22 (and the sound quality of the acoustic signal z (t) is lowered). In the first embodiment, since the peak frequency of the reverberation spectrum YR (k, m) of the reverberation component of the observation signal x (t) is specified as the reverberation frequency FR, the initial sound component of the observation signal x (t) is specified. Is reduced compared to the proportionality. Therefore, according to the first embodiment, there is an advantage that howling can be prevented while suppressing deterioration in sound quality due to band suppression.

Next, a specific configuration of the analysis processing unit 34 will be described with reference to FIG. As shown in FIG. 3, the analysis processing unit 34 of the first embodiment includes an index value calculation unit 42 </ b> A and a dominance calculation unit 44. The index value calculation unit 42A sequentially calculates the first index value Q1 (k, m) and the second index value Q2 (k, m) corresponding to the observation signal x (t). Specifically, the index value calculation unit 42A includes a first smoothing unit 51 and a second smoothing unit 52. The first smoothing unit 51 smoothes the time series of the power | X (k, m) | ² of the observation signal x (t) to obtain the first index value Q1 (k, m) of each frequency for each unit period. Calculate sequentially. Similarly, the second smoothing unit 52 smoothes the time series of the power | X (k, m) | ² of the observation signal x (t) to obtain the second index value Q2 (k, m) of each frequency. Calculate sequentially for each unit period.

The first index value Q1 (k, m) is defined in the following formula (1A), and the first index value Q1 (k, m) is within a first period composed of successive N1 (N1 is a natural number of 2 or more) unit periods. It is a moving average (simple moving average) of power | X (k, m) | ² . The first period is a set of N1 unit periods, for example, with the m-th unit period at the end. On the other hand, the second index value Q2 (k, m) is a second period composed of N2 (N2 is a natural number greater than or equal to 2) unit periods as defined by the following formula (1B). X (k, m) | | power in the inner a ^second moving average. The second period is a set of N2 unit periods, for example, with the m-th unit period at the end. As understood from the above description, the first smoothing unit 51 and the second smoothing unit 52 correspond to an FIR (finite impulse response) type low-pass filter.

  The number N2 of unit periods added to the calculation of the second index value Q2 (k, m) exceeds the number N1 of unit periods added to the calculation of the first index value Q1 (k, m) (N2> N1). ). That is, the second period is longer than the first period. For example, the first period is set to a time of about 100 milliseconds to 300 milliseconds, and the second period is set to a time of about 300 milliseconds to 600 milliseconds. Therefore, the time constant τ2 for smoothing by the second smoothing unit 52 exceeds the time constant τ1 for smoothing by the first smoothing unit 51 (τ2> τ1). Assuming that the first smoothing unit 51 and the second smoothing unit 52 are realized by a low-pass filter, it can be said that the cutoff frequency of the second smoothing unit 52 is lower than the cutoff frequency of the first smoothing unit 51.

Part (B) of FIG. 4 is a graph of the time change of the first index value Q1 (k, m) and the second index value Q2 (k, m) calculated for an arbitrary frequency of the observation signal x (t). is there. When the room impulse response (RIR) in which the power | X (k, m) | ² (power density) exponentially decays is supplied to the sound processing apparatus 100 as the observation signal x (t) as shown in part (A) of FIG. The first index value Q1 (k, m) and the second index value Q2 (k, m) are shown in part (B) of FIG.

As understood from the part (B) of FIG. 4, the first index value Q1 (k, m) and the second index value Q2 (k, m) are the power | X (k, m) | Follows ² and changes over time. However, since the time constant τ2 of smoothing by the second smoothing unit 52 exceeds the time constant τ1 of smoothing by the first smoothing unit 51, the second index value Q2 (k, m) is the first index value Q1 (k , m) follows the time change of the power | X (k, m) | ² of the observation signal x (t) with lower followability (change rate). Specifically, as shown in part (B) of FIG. 4, the first index value Q1 (k, m) is the second index value Q2 (k, m,) in the section immediately after the time t0 of the start of the indoor impulse response. Increase at a rate of change above m). The first index value Q1 (k, m) and the second index value Q2 (k, m) reach peaks at different times on the time axis, and the first index value Q1 (k, m) 2 Decreases at a rate of change exceeding the index value Q2 (k, m).

  As described above, since the first index value Q1 (k, m) and the second index value Q2 (k, m) change at different rates, the first index value Q1 (k, m) and the second index value The magnitude of the value Q2 (k, m) is inverted at a specific time point tx on the time axis. That is, the first index value Q1 (k, m) exceeds the second index value Q2 (k, m) in the section SA from the time t0 to the time tx, and the second index value Q2 (k in the section SB after the time tx. , m) exceeds the first index value Q1 (k, m). The section SA corresponds to a section where the initial sound component (direct sound and initial reflected sound) of the room impulse response exists, and the section SB corresponds to a section where the reverberation component (rear reverberation sound) of the room impulse response exists.

  The dominance calculation unit 44 in FIG. 3 has a reverberation dominance Gr (k, k, m) according to the first index value Q1 (k, m) and the second index value Q2 (k, m) calculated by the index value calculation unit 42A. m) is sequentially calculated for each frequency for each unit period. The dominance calculation unit 44 of the first embodiment includes a ratio calculation unit 62, a first processing unit 64, and a second processing unit 66.

The ratio calculator 62 calculates the ratio R (k, m) between the first index value Q1 (k, m) and the second index value Q2 (k, m). Specifically, the ratio calculation unit 62 expresses the ratio R (k, m) of the first index value Q1 (k, m) to the second index value Q2 (k, m) as expressed by the following formula (2). m) is calculated for each unit period.

  The first processing unit 64 of FIG. 3 sequentially calculates the initial sound dominance Ge (k, m) for each frequency for each unit period in accordance with the ratio R (k, m) calculated by the ratio calculation unit 62. In contrast to the reverberation dominance Gr (k, m), which corresponds to the index of the ratio of the reverberation component, the initial sound dominance Ge (k, m) is the initial sound (direct sound and direct sound) of the observed signal x (t). It is a variable that serves as an index of the ratio of the initial reflected sound. Accordingly, generally, the initial sound dominance Ge (k, m) increases as the intensity of the reverberation component in the spectrum X (k, m) of the observed signal x (t) decreases relative to the intensity of the initial sound component. Set to a numeric value. In other words, the initial sound dominance Ge (k, m) can be rephrased as the dominance of the initial sound component in the spectrum X (k, m) of the observation signal x (t).

  The first processing unit 64 of the first embodiment compares the ratio R (k, m) calculated by the ratio calculation unit 62 with the predetermined value Gmax and the predetermined value Gmin, and the initial sound dominance Ge (k, k, m) is calculated for each unit period. The predetermined value Gmax and the predetermined value Gmin are threshold values that are set in advance in accordance with, for example, an instruction from the user and compared with the ratio R (k, m). In the first embodiment, a case where the predetermined value Gmax is set to 1 is exemplified. The predetermined value Gmin is set to a numerical value lower than the predetermined value Gmax (a numerical value in the range of 0 or more and less than 1).

Specifically, the first processing unit 64 performs the calculation of the following mathematical formula (3). First, when the ratio R (k, m) exceeds the predetermined value Gmax (Gmax = 1) (R (k, m) ≧ Gmax), the first processing unit 64 sets the predetermined value Gmax to the initial sound dominance Ge. Set as (k, m). Second, when the ratio R (k, m) is lower than the predetermined value Gmin (R (k, m) ≦ Gmin), the first processing unit 64 sets the predetermined value Gmin to the initial sound dominance Ge (k, m). Set as. Third, when the ratio R (k, m) is a numerical value between the predetermined value Gmax and the predetermined value Gmin (Gmin <R (k, m) <Gmax), the first processing unit 64 uses the ratio R ( k, m) is set as the initial sound dominance Ge (k, m).

  The change of the initial sound dominance Ge (k, m) when the first index value Q1 (k, m) and the second index value Q2 (k, m) change as shown in the part (B) of FIG. This is illustrated in part 4 (C). As understood from the part (C) of FIG. 4, the initial sound when the first index value Q1 (k, m) exceeds the second index value Q2 (k, m) (section SA) is roughly shown. The dominance degree Ge (k, m) is based on the initial sound dominance degree Ge (k, m) when the first index value Q1 (k, m) is lower than the second index value Q2 (k, m) (section SB). Is also a large number. Specifically, the ratio R (k, m) exceeds the predetermined value Gmax (Gmax = 1) in the section SA where the first index value Q1 (k, m) exceeds the second index value Q2 (k, m). Thus, the initial sound dominance Ge (k, m) is maintained at a predetermined value Gmax. In the section SB in which the ratio R (k, m) exceeds the predetermined value Gmin in the section SB where the first index value Q1 (k, m) is lower than the second index value Q2 (k, m), the initial sound dominance level Ge (k, m) is set to the ratio R (k, m) and decreases with time. In the section SB2 in which the ratio R (k, m) is below the predetermined value Gmin in the section SB, the initial sound dominance Ge (k, m) is maintained at the predetermined value Gmin.

  That is, the initial sound dominance Ge (k, m) calculated by the first processing unit 64 is set to a predetermined value (maximum value) Gmax in the section SA where the direct sound and the initial reflected sound exist, and a reverberation component exists. In the section SB, it decreases with time to a predetermined value (minimum value) Gmin. Therefore, when the spectrum X (k, m) of the observed signal x (t) is multiplied by the initial sound dominance Ge (k, m), the reverberation component of the observed signal x (t) is suppressed (direct sound or initial reflected sound). Is generated). That is, when the initial sound dominance Ge (k, m) is applied as a gain for the spectrum X (k, m) of the observation signal x (t), the reverberation component of the observation signal x (t) is suppressed. Used as a coefficient (adjustment value for reverberation suppression).

  The second processing unit 66 sequentially calculates the reverberation dominance Gr (k, m) corresponding to the initial sound dominance Ge (k, m) calculated by the first processing unit 64 for each frequency for each unit period. The reverberation dominance Gr (k, m) is calculated so that the reverberation dominance Gr (k, m) decreases as the initial sound dominance Ge (k, m) increases. Specifically, the second processing unit 66 subtracts the initial sound dominance Ge (k, m) calculated by Expression (3) from a predetermined value (1 in the following example), thereby reverberation dominance Gr ( k, m) is calculated (Gr (k, m) = 1-Ge (k, m)). Therefore, the reverberation dominance Gr (k, m) is maintained at zero in the section SA where the initial sound component exists, and increases with time to a predetermined value (1-Gmin) in the section SB where the reverberation component exists. That is, when the first index value Q1 (k, m) exceeds the second index value Q2 (k, m) (section SA), the reverberation dominance Gr (k, m) is the first index value Q1 (k, m When m) is lower than the second index value Q2 (k, m) (section SB), the reverberation dominance Gr (k, m) is smaller. Therefore, the reverberation dominance Gr (k, m) calculated by the second processing unit 66 is applied to the calculation of the reverberation extraction unit 36 (YR (k, m) = Gr (k, m) X (k, m)). Thus, the reverberation spectrum YR (k, m) of the reverberation component is calculated.

  As described above, in the first embodiment, the ratio R () of the first index value Q1 (k, m) and the second index value Q2 (k, m) following the time change of the observation signal x (t). Since the reverberation dominance Gr (k, m) is calculated according to k, m), there is an advantage that the reverberation component of the observation signal x (t) can be extracted by simple processing.

Second Embodiment
A second embodiment of the present invention will be described below. In addition, about the element in which an effect | action and a function are equivalent to 1st Embodiment in each form illustrated below, the code | symbol referred by description of 1st Embodiment is diverted, and each detailed description is abbreviate | omitted suitably.

  The acoustic processing apparatus 100 of the second embodiment has a configuration in which the reverberation analysis unit 24A of the first embodiment is replaced with a reverberation analysis unit 24B of FIG. As shown in FIG. 5, the reverberation analysis unit 24B of the second embodiment includes a frequency analysis unit 32, an analysis processing unit 34, and a frequency specifying unit 38. That is, the reverberation extraction unit 36 of the first embodiment is omitted. The operations and configurations of the frequency analysis unit 32 and the analysis processing unit 34 are the same as those in the first embodiment. That is, the frequency analysis unit 32 generates a spectrum X (k, m) of the observation signal x (t) for each unit period, and the analysis processing unit 34 reverberation dominance Gr ( k, m) are calculated sequentially for each frequency for each unit period.

  Since the reverberation dominance Gr (k, m) corresponds to the ratio of the reverberation component in the observed signal x (t), the reverberation dominance Gr (k, m) is a large value at the frequency where the reverberation component is abundant. Tend to be. Therefore, the frequency specifying unit 38 of the second embodiment specifies the peak frequency of the reverberation dominance Gr (k, m) on the frequency axis as the reverberation frequency FR (m). Specifically, the frequency specifying unit 38 specifies the frequency at which the reverberation dominance Gr (k, m) has the maximum value as the reverberation frequency FR (m). In order to suppress instantaneous fluctuations of the reverberation frequency FR (m) due to noise or the like, the reverberation dominance Gr (k, m) is smoothed in the time axis direction and then the reverberation frequency FR for each unit period. It is also possible to specify (m). The operation of the band suppression unit 22 to suppress the band component including the reverberation frequency FR (m) specified by the frequency specifying unit 38 in the observed signal x (t) is the same as that of the first embodiment.

  In the second embodiment, the same effect as in the first embodiment is realized. Further, in the second embodiment, since extraction of reverberation components is unnecessary, there is an advantage that the configuration and operation of the acoustic processing measure are simplified as compared with the first embodiment.

<Third Embodiment>
FIG. 6 is a block diagram of the sound processing apparatus 100 according to the third embodiment. As shown in FIG. 6, the sound processing apparatus 100 according to the third embodiment is configured by adding a dereverberation unit 72 to the same elements (band suppression unit 22, reverberation analysis unit 24A, amplification unit 26) as in the first embodiment. It is. The reverberation suppression unit 72 generates the observation signal v (t) by suppressing the reverberation component in the observation signal x (t) generated by the sound collection device 12. The band suppression unit 22 suppresses band components including the reverberation frequency FR (m) in the observation signal v (t) after the reverberation suppression by the reverberation suppression unit 72.

  Similarly to the example of FIG. 3, the analysis processing unit 34 of the reverberation analysis unit 24A of the third embodiment includes a first processing unit 64 that calculates an initial sound dominance Ge (k, m), and a reverberation dominance Gr (k , m) and a second processing unit 66 that calculates m). 6 applies the initial sound dominance Ge (k, m) calculated by the first processing unit 64 to the observation signal x (t) to generate the observation signal v (t). Specifically, the reverberation suppression unit 72 multiplies the spectrum X (k, m) of the observed signal x (t) by the initial sound dominance Ge (k, m) to thereby obtain the spectrum V of the observed signal v (t). (k, m) is generated (V (k, m) = Ge (k, m) X (k, m)), and the spectrum V (k, m) is converted into the time domain by, for example, short-time inverse Fourier transform. Thus, the observation signal v (t) is generated.

  As described above, the initial sound dominance Ge (k, m) is set to the predetermined value Gmax in the section SA where the initial sound component exists predominantly, and the time lapses until the predetermined value Gmin in the section SB where the reverberation component exists predominantly. Decrease. Therefore, the observation signal v (t) processed by the reverberation suppressing unit 72 is an acoustic signal in which the initial sound component is emphasized (the reverberation component is suppressed). That is, the initial sound dominance Ge (k, m) corresponds to a gain with respect to the spectrum X (k, m) of the observation signal x (t), and is a coefficient for suppressing the reverberation component of the observation signal x (t). Used.

  In the third embodiment, the same effect as in the first embodiment is realized. In the first embodiment in which only the band component of the reverberation frequency FR (m) is suppressed in the observation signal x (t), the band components other than the reverberation frequency FR (m) are maintained and increase with time. It may cause howling. On the other hand, in the third embodiment, in addition to the suppression of the band component of the reverberation frequency FR (m) by the band suppression unit 22, the reverberation component is suppressed by the reverberation suppression unit 72. Therefore, as compared with the first embodiment that executes only suppression of the band component of the reverberation frequency FR (m), it is possible to suppress an increase in howling over time for band components other than the reverberation frequency FR (m). is there. In the third embodiment, the initial sound dominance Ge (k, m) calculated by the first processing unit 64 is calculated from the reverberation dominance Gr (k, m) by the second processing unit 66 and the dereverberation suppression unit 72. Therefore, there is an advantage that the processing is simplified compared with a configuration in which the calculation of the reverberation dominance Gr (k, m) and the suppression of the reverberation component are performed independently. . However, a configuration in which the reverberation suppression unit 72 suppresses the reverberation component of the observation signal x (t) independently of the calculation of the initial sound dominance Ge (k, m) by the first processing unit 64 may be employed.

  It is also possible to apply the third embodiment to the second embodiment. Further, the order of the reverberation suppressing unit 72 and the band suppressing unit 22 can be reversed. Specifically, the reverberation suppression unit 72 suppresses the reverberation component of the acoustic signal z (t) processed by the band suppression unit 22 (the band suppression unit 22 processes the observation signal before the processing by the reverberation suppression unit 72). Configuration) can also be employed.

<Fourth embodiment>
FIG. 7 is a block diagram of the sound processing apparatus 100 according to the fourth embodiment. As shown in FIG. 7, the sound processing apparatus 100 according to the fourth embodiment has a configuration in which a delay unit 74 is added to the same elements (band suppression unit 22, reverberation analysis unit 24A, amplification unit 26) as in the first embodiment. is there. The delay unit 74 delays the observation signal x (t) generated by the sound collection device 12. The reverberation analysis unit 24A processes the observation signal x (t) before the delay by the delay unit 74, and the band suppression unit 22 processes the observation signal x (t) after the delay by the delay unit 74.

  The processing by the reverberation analysis unit 24A (calculation of the reverberation dominance Gr (k, m) and detection of the reverberation frequency FR (m)) is performed in the frequency domain, whereas the band suppression by the band suppression unit 22 is performed in the time domain. Executed. Therefore, in the first embodiment, the calculation of the reverberation frequency FR (m) by the analysis processing unit 34 is delayed with respect to the operation of the band suppression unit 22 by a time corresponding to the window size of the short-time Fourier transform by the frequency analysis unit 32. To do. In order to eliminate the delay described above, the delay unit 74 of the fourth embodiment delays the observation signal x (t) by a delay amount corresponding to the window size of the short-time Fourier transform by the frequency analysis unit 32. Therefore, at the stage where the band suppression unit 22 processes a specific time point of the observation signal x (t), the reverberation frequency FR () specified by the analysis processing unit 34 for the unit period corresponding to that time point in the observation signal x (t). m) is instructed to the band suppression unit 22.

  In the fourth embodiment, the same effect as in the first embodiment is realized. In the fourth embodiment, the reverberation frequency FR (m) specified for each unit period of the observation signal x (t) is applied to band suppression at the time corresponding to the unit period of the observation signal x (t). Therefore, for example, even if the acoustic characteristics of the observation signal x (t) (reverberation dominance Gr (k, m) and initial sound dominance Ge (k, m)) frequently change in a short time, There is an advantage that howling can be effectively prevented. Note that the fourth embodiment can also be applied to the second embodiment and the third embodiment.

<Fifth Embodiment>
FIG. 8 is a block diagram of the analysis processing unit 34 in the fifth embodiment. The analysis processing unit 34 of the fifth embodiment has a configuration in which the index value calculation unit 42A of the first embodiment illustrated in FIG. 3 is replaced with an index value calculation unit 42B. The index value calculator 42B is an element that sequentially calculates the first index value Q1 (k, m) and the second index value Q2 (k, m) for each unit period. The first smoother 51 and the second smoother A unit 52 and a delay unit 54 are included. The configuration and operation of the dominance calculation unit 44 are the same as those in the first embodiment.

Similarly to the first embodiment, the first smoothing unit 51 smoothes the time series of the power | X (k, m) | ² of the observation signal x (t) to thereby provide the first index value Q1 (k, m ) For each unit period. The delay unit 54 is a storage circuit that delays the spectrum X (k, m) of the observation signal x (t) by a time corresponding to d units (d is a natural number). The second smoothing unit 52 smoothes the time series of the power | X (k, m) | ² of the spectrum X (k, m) delayed by the delay unit 54 to thereby generate the second index value Q2 (k, m ) For each unit period. However, in the second embodiment, the time constant τ 2 for smoothing by the second smoothing unit 52 is equivalent to the time constant τ 1 for smoothing by the first smoothing unit 51 (τ 2 = τ 1). Therefore, the time change of the second index value Q2 (k, m) has a relationship in which the time change of the first index value Q1 (k, m) is delayed by d units (Q2 (k, m). ) = Q1 (k, md)). Note that the time constant τ1 of smoothing by the first smoothing unit 51 and the time constant τ2 of smoothing by the second smoothing unit 52 can be made different. Further, a configuration in which the second index value Q2 (k, m) is calculated by delaying the first index value Q1 (k, m) calculated by the first smoothing unit 51 (a configuration in which the second smoothing unit 52 is omitted). Can also be employed.

  The part (B) of FIG. 9 supplies the indoor impulse response (part (A) of FIG. 9) similar to the part (A) of FIG. 4 to the sound processing apparatus 100 of the second embodiment as the observation signal x (t). It is a graph of the time change of the 1st index value Q1 (k, m) and the 2nd index value Q2 (k, m) at the time of doing.

As understood from the part (B) of FIG. 9, the first index value Q1 (k, m) and the second index value Q2 (k, m) have the same time change mode (waveform). The time change of the two index values Q2 (k, m) is delayed by d unit times with respect to the time change of the first index value Q1 (k, m). That is, the second index value Q2 (k, m) is reduced to the power | X (k, m) | ² of the observation signal x (t) with lower followability than the first index value Q1 (k, m). Follow. Therefore, as in the first embodiment, the magnitudes of the first index value Q1 (k, m) and the second index value Q2 (k, m) are inverted at a specific time point tx on the time axis. That is, the first index value Q1 (k, m) exceeds the second index value Q2 (k, m) in the section SA up to the time tx, and the second index value Q2 (k, m) in the section SB after the time tx. Exceeds the first index value Q1 (k, m).

  Calculation of the ratio R (k, m) by the ratio calculation unit 62 (formula (2)), calculation of the initial sound dominance Ge (k, m) by the first processing unit 64, and reverberation dominance Gr by the second processing unit 66 The calculation of (k, m) is the same as in the first embodiment. Therefore, as shown in part (C) of FIG. 9, the initial sound dominance Ge (k, m) is set to a predetermined value Gmax in the section SA where the direct sound and the initial reflected sound exist, and the rear reverberation sound is In the existing section SB, it decreases with time to a predetermined value Gmin. Similar to the first embodiment, the reverberation frequency FR (m) is specified using the reverberation dominance degree Gr (k, m) corresponding to the initial sound dominance degree Ge (k, m). In the fifth embodiment, the same effect as in the first embodiment is realized. It is possible to apply the fifth embodiment from the second embodiment to the fourth embodiment.

<Modification>
Each of the above-described embodiments can be variously modified. Specific modifications are exemplified below. Two or more aspects arbitrarily selected from the following examples can be appropriately combined.

(1) In each of the above-described embodiments, the simple moving average of the power | X (k, m) | ² of the observation signal x (t) is calculated as the first index value Q1 (k, m) and the second index value Q2 (k, Although calculated as m), the calculation method of the first index value Q1 (k, m) and the second index value Q2 (k, m) is not limited to the above examples. For example, as expressed by the following formula (4A) and formula (4B), the exponential average (exponential moving average) of the power | X (k, m) | ² of the observation signal x (t) is the first index value. It is also possible to calculate as Q1 (k, m) and the second index value Q2 (k, m).

That is, the first smoothing unit 51 and the second smoothing unit 52 correspond to an IIR (infinite impulse response) type low-pass filter. Symbol α1 in equation (4A) and symbol α2 in equation (4B) are smoothing coefficients (forgetting coefficients). Specifically, the smoothing coefficient α1 means the weight of the current power | X (k, m) | ² with respect to the past first index value Q1 (k, m−1), and the smoothing coefficient α2 is This means the weight of the current power | X (k, m) | ² with respect to the past second index value Q2 (k, m-1). The smoothing coefficient α2 is set to a numerical value lower than the smoothing coefficient α1 (α2 <α1). Therefore, as in the first embodiment, the time constant τ2 for smoothing by the second smoothing unit 52 exceeds the time constant τ1 for smoothing by the first smoothing unit 51 (τ2> τ1). That is, the second index value Q2 (k, m) is reduced to the power | X (k, m) | ² of the observation signal x (t) with lower followability than the first index value Q1 (k, m). Follow.

Further, as expressed by the following formulas (5A) and (5B), the weighted moving average of the power | X (k, m) | ² of the observation signal x (t) is expressed as the first index value Q1 (k, m) and the second index value Q2 (k, m) can also be calculated. Symbol w1 (i) in equation (5A) and symbol w2 (i) in equation (5B) mean weight values for the i-th unit period located in front of the m-th unit period. The condition that the second period is longer than the first period (N2> N1) is the same as the above example.

(2) In each of the above embodiments, the frequency specifying unit 38 specifies one reverberation frequency FR (m) for each unit period. However, it is also possible to specify a plurality of reverberation frequencies FR (m) for each unit period. It is. For example, the frequency specifying unit 38 selects a predetermined number of peaks in descending order of the intensity of the reverberation spectrum YR (k, m), and specifies the frequency of each peak as the reverberation frequency FR (m). The band suppression unit 22 suppresses the band component of each reverberation frequency FR (m) in the observation signal x (t). For example, an element in which a total number of notch filters corresponding to the number of reverberation frequencies FR (m) are cascaded is employed as the band suppression unit 22.

初期音優勢度Ｇe(k,m)を使用する構成（例えば第３実施形態）において残響優勢度Ｇr(k,m)を数式(3A)で算定する場合、初期音優勢度Ｇe(k,m)を前掲の数式(3)で算定する構成（すなわち、初期音優勢度Ｇe(k,m)と残響優勢度Ｇr(k,m)とを並列に算定する構成）や、数式(3A)で算定された残響優勢度Ｇr(k,m)を１から減算することで初期音優勢度Ｇe(k,m)（Ｇe(k,m)＝１−Ｇr(k,m)）を算定する構成（すなわち、第１処理部６４と第２処理部６６とを入替えた構成）も採用され得る。 (3) The calculation method of the reverberation dominance degree Gr (k, m) and the initial sound dominance degree Ge (k, m) is arbitrary. For example, as will be understood from the description of each embodiment described above, the reverberation dominance Gr (k, m) can be calculated by the following equation (3A).

When the reverberation dominance Gr (k, m) is calculated by the formula (3A) in the configuration using the initial sound dominance Ge (k, m) (for example, the third embodiment), the initial sound dominance Ge (k, m ) With the above formula (3) (that is, a configuration in which the initial sound dominance Ge (k, m) and the reverberation dominance Gr (k, m) are calculated in parallel) and the formula (3A) Subtracting the calculated reverberation dominance Gr (k, m) from 1 to calculate the initial sound dominance Ge (k, m) (Ge (k, m) = 1-Gr (k, m)) (In other words, a configuration in which the first processing unit 64 and the second processing unit 66 are interchanged) may be employed.

  Further, for example, reverberation predominance is obtained by a predetermined calculation using the first index value Q1 (k, m) and the second index value Q2 (k, m) as a variable or a predetermined calculation using the ratio R (k, m) as a variable. A configuration for calculating the degree Gr (k, m) and the initial sound dominance Ge (k, m) can also be adopted. In each of the above-described embodiments, the reverberation dominance Gr (k, m) is calculated according to the ratio R (k, m) of the first index value Q1 (k, m) to the second index value Q2 (k, m). However, for example, the ratio R (k, m) of the second index value Q2 (k, m) to the first index value Q1 (k, m) is applied to the calculation of the equation (3), whereby the reverberation dominance Gr ( It is also possible to calculate k, m) directly (that is, without calculating the initial sound dominance Ge (k, m)). That is, the calculation of the initial sound dominance Ge (k, m) can be omitted. As understood from the above description, the dominance calculation unit 44 reverberation according to the ratio of the reverberation component of the observation signal x (t) (ratio of the reverberation component intensity to the signal intensity of the observation signal x (t)). It is included as an element for calculating the dominance Gr (k, m).

(4) Although the band suppression by the band suppression unit 22 is performed with the frequency at which the intensity of the reverberation spectrum YR (k, m) is maximum as the reverberation frequency FR (m), the reverberation component is almost present in the observed signal x (t). If not, the band suppression by the band suppression unit 22 can be stopped. For example, when the maximum value of the intensity of the reverberation spectrum YR (k, m) is lower than a predetermined threshold TH (that is, when there is almost no reverberation component), a configuration is adopted in which the suppression of the band component by the band suppression unit 22 is stopped. Is done. The threshold value TH is set to a fixed value. According to the above configuration, there is an advantage that deterioration of sound quality due to band suppression can be suppressed in a state where the reverberation component is hardly present in the observation signal x (t). It is also possible to variably control the threshold value TH. For example, a configuration is adopted in which a numerical value obtained by adding or subtracting a predetermined value to the average value of the reverberation spectrum YR (k, m) is set as the threshold TH.

(5) In each of the above-described embodiments, the notch filter is exemplified as the band suppression unit 22, but the configuration and operation of the band suppression unit 22 are arbitrary. For example, an equalizer (parametric equalizer) that suppresses the band component of the reverberation frequency FR (m) in the observation signal x (t) is used as the band suppression unit 22. Further, the band component of the reverberation frequency FR (m) from the spectrum X (k, m) of the observed signal x (t) (for example, the band component including the reverberation frequency FR (m) of the reverberation spectrum YR (k, m) (Predetermined band component prepared in FIG. 4) is subtracted (spectrum subtraction), or the band component including the reverberation frequency FR (m) of the spectrum X (k, m) of the observation signal x (t) is less than one. A configuration of multiplying the value (spectral gain) can also be adopted. It is also possible to use a plurality of methods for suppressing the band component of the observation signal x (t).

(6) In each of the above embodiments, the time constant τ1 for smoothing by the first smoothing unit 51 and the time constant τ2 for smoothing by the second smoothing unit 52 are made common over a plurality of frequencies. It is also possible to individually set the time constant τ2 for each frequency (for each band). It is also possible to change one or both of the time constant τ1 and the time constant τ2 over time.

(7) In the first embodiment, one frequency that maximizes the reverberation spectrum YR (k, m) is specified as the reverberation frequency FR (m), and in the second embodiment, the reverberation dominance Gr (k, m) is Although one maximum frequency is specified as the reverberation frequency FR (m), the method for specifying the reverberation frequency FR (m) may be appropriately changed. For example, the frequency at which the amount of increase over time of the reverberation spectrum YR (k, m) (the amount of increase in intensity from the reverberation spectrum YR (k, m) in the previous unit period) is the maximum is the reverberation frequency FR (m). A configuration for specifying the frequency at which the amount of increase over time of the reverberation dominance GR (k, m) becomes maximum can be adopted as the reverberation frequency FR (m).

(8) In each of the above embodiments, the first index value Q1 (k, m) and the second index value are smoothed by smoothing the time series of the power | X (k, m) | ² of the observation signal x (t). Although Q2 (k, m) is calculated, the object of smoothing by the first smoothing unit 51 and the second smoothing unit 52 is not limited to the power | X (k, m) | ² . For example, by smoothing the amplitude | X (k, m) | of the observation signal x (t) and the fourth power of the amplitude | X (k, m) | ⁴ , the first index value Q1 (k, m) A configuration for calculating the two index values Q2 (k, m) can also be adopted. That is, the first smoothing unit 51 and the second smoothing unit 52 in the above-described embodiments are included as elements that smooth the time series of the signal strength of the observation signal x (t), and the signal strength is the observation signal x (t ) Power | X (k, m) | ² , the amplitude | X (k, m) | and the fourth power of amplitude | X (k, m) | ⁴ are included. In each of the above-described embodiments, the reverberation extraction unit 36 causes the reverberation dominance Gr (k, m) to act on the spectrum X (k, m) of the observation signal x (t), and the reverberation suppression unit 72 performs the initial sound dominance Ge (k, m) spectra X (k, m) of the observed signal x (t) is allowed to act on, for example, the power of the observed signal x (t) | X (k , m) | 2 in reverberation dominance Gr (k, m) and initial sound dominance Ge (k, m) can be applied.

(9) In each of the above-described embodiments, the acoustic processing device 100 that functions as a howling suppression device is exemplified by the configuration including the band suppression unit 22 and the reverberation analysis unit 24 (24A, 24B), but the reverberation analysis unit 24 is provided. The present invention can also be implemented as a howling frequency identification device that identifies the reverberation frequency FR (m) by a configuration (for example, a configuration in which the band suppression unit 22 and the amplification unit 26 are omitted from the above-described embodiments).

DESCRIPTION OF SYMBOLS 100 ... Sound processing apparatus, 10 ... Loudspeaker, 12 ... Sound collection apparatus, 14 ... Sound emission apparatus, 22 ... Band suppression part, 24A, 24B ... Reverberation analysis part, 26 ... Amplification part, 32 ...... Frequency analysis unit 34... Analysis processing unit 36... Reverberation extraction unit 38... Frequency identification unit 42 A, 42 B .. Index value calculation unit 44 .. Dominance calculation unit 51. Smoothing unit 52... Second smoothing unit 54... Delay unit 62... Ratio calculation unit 64... First processing unit 66. Delay part.

Claims (6)

Reverberation analysis means for estimating the frequency of the peak of the spectrum of the reverberation component included in the observation signal as the reverberation frequency for each frame on the time axis;
Band processing means for suppressing band components including the reverberation frequency in the observation signal. The reverberation analysis means includes
Analysis processing means for calculating the reverberation dominance according to the ratio of the reverberation component of the observation signal for each frequency;
Reverberation extraction means for extracting a reverberation component by applying the reverberation dominance of each frequency to the observation signal;
The sound processing apparatus according to claim 1, further comprising: a frequency specifying unit that specifies a peak frequency of a reverberation component spectrum extracted by the reverberation extracting unit as the reverberation frequency. Reverberation analysis means for estimating the reverberation frequency as the frequency at which the reverberation component of the observed signal exists
Band suppression means for suppressing a band component including the reverberation frequency in the observation signal,
The reverberation analysis means extracts reverberation components by applying reverberation dominance of each frequency to the observation signal, and analysis processing means for calculating the reverberation dominance according to the ratio of the reverberation component of the observation signal for each frequency. Reverberation extracting means, and frequency specifying means for specifying the peak frequency of the spectrum of the reverberation component extracted by the reverberation extracting means as a reverberation frequency,
The analysis processing means calculates a first index value that follows the time change of the observation signal and a second index value that follows the time change of the observation signal with lower followability than the first index value. An acoustic processing apparatus, comprising: index value calculating means; and dominance degree calculating means for calculating the reverberation dominance degree according to a difference between the first index value and the second index value. Comprising reverberation suppression means for suppressing the reverberation component of the observation signal;
The acoustic processing device according to any one of claims 1 to 3, wherein the band suppression unit suppresses a band component including the reverberation frequency in an observation signal before or after processing by the reverberation suppression unit. Comprising delay means for delaying the observation signal;
The reverberation analysis means uses the observation signal before processing by the delay means as a processing target,
The acoustic processing apparatus according to claim 1, wherein the band suppression unit targets an observation signal processed by the delay unit. Computer
Estimate the frequency of the peak of the reverberation component spectrum included in the observed signal for each frame on the time axis as the reverberation frequency,
An acoustic processing method for suppressing a band component including the reverberation frequency in an observation signal.

Images