JP2016177004A - Signal processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To execute control so that auditory impressions at the switching timing of sound contents can be natural without losing the acoustic feature of a plurality of continuously reproduced sound contents.SOLUTION: A signal processor includes analysis means, changing amount calculation means, and changing means. The analysis means analyzes the acoustic feature of each sound content in a section before/after the switching timing of a plurality of continuously reproduced sound contents. The changing amount calculation means calculates the changing amount of the acoustic feature amount of each sound content in the section before/after the switching timing based on an acoustic feature amount indicating the acoustic feature of the sound content in the section before the switching timing and an acoustic feature amount indicating the acoustic feature of the sound content in the section after the switching timing. The changing means executes processing for the sound content in the section before/after the switching timing according to the changing amount calculated by the changing amount calculation means.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a signal processing technique for a sound signal, and more particularly to a technique for adjusting acoustic characteristics such as volume, frequency characteristics, and reverberation characteristics.

  In recent years, various kinds of music content such as music clips and recorded data of concerts are available through the Internet. When using these music contents, a plurality of music contents selected according to the user's preference are rearranged so that they can be continuously played on a playback device such as a portable music player, and new music contents are edited. There is.

  When a plurality of music contents are continuously played back, when the music contents are switched, a sense of incongruity or an auditory gap (hereinafter referred to as an auditory gap) may occur. As a cause of occurrence of an audible gap or the like, there is a difference in acoustic characteristics (hereinafter referred to as acoustic characteristics) between the music content being reproduced and the subsequent music content. For example, if there is a difference between the volume of the music content being played and the volume of the subsequent music content, the volume difference is experienced as an audible gap. Further, when the music content being reproduced is recorded in the studio and the subsequent music content is recorded live, the difference in the reverberation characteristics between the two is experienced as an audible gap or the like. Hereinafter, a physical quantity representing an acoustic feature is referred to as an acoustic feature quantity. Specific examples of the acoustic feature amount include volume and amount and quality of reverberant sound.

  In order to prevent an audible gap or the like from occurring, the difference between the acoustic features of the music content being played and the subsequent music content is alleviated, that is, the difference between the acoustic features of the two music contents is eliminated (or reduced). It is necessary to. An example of a technique that enables this is the technique disclosed in Patent Document 1. In the technique disclosed in Patent Document 1, a plurality of audio data is analyzed in advance to design a target characteristic for an acoustic feature, and each audio data is corrected so that the acoustic feature approaches the target characteristic. For example, when the acoustic feature is a volume, the arithmetic average of the volume of each audio data is set as the target characteristic, and the volume of each audio data is corrected so as to approach the target characteristic.

JP 2003-273678 A

Tadanobu Kondo, Internet, [online], <URL: http://reverb2014.dereverberation.com/workshop/index.html> K. Lebart, et al., Acta acustica, ACUSTICA, Vol. 87 (2001), pp. 359-366 Jim Y.C. Wen, et al., Acoustics, Speech and SignalProcessing, 2008.ICASSP 2008. March 31 2008-April 4 2008, pp.329-332 Keisuke Kinoshita, et al., IEEE TRANSACTION ON ON AUDIO, SPEECH ANDLANUAGE PROCESSING, VOL.17, NO.4, MAY 2009, pp.1-12

  However, the technique disclosed in Patent Document 1 has the following problems. First, it is necessary to perform a large-scale process in advance in order to determine the target characteristic. Secondly, the way of listening through the entire music content changes due to the correction, and the original characteristics of each music content (for example, the quality of a live sound source) are impaired. Therefore, it is not possible to control the audibility before and after the switching of the music contents without impairing the acoustic characteristics of each of the plurality of music contents to be reproduced. The same problem also occurs when a plurality of sound contents other than music contents such as environmental sounds and text-to-speech sounds, masker sounds, etc. are continuously played back.

  The present invention has been made in view of the above problems, and it is natural to have an audible impression at the switching timing of these sound contents without deteriorating the acoustic characteristics of a plurality of sound contents to be reproduced continuously. It is an object of the present invention to provide a technique that enables control to be performed.

  In order to solve the above-described problems, the present invention provides an analysis means for analyzing acoustic characteristics of each sound content in a section before and after the switching timing of a plurality of sound contents to be played back continuously, and before the switching timing. Based on the acoustic feature amount representing the acoustic feature of the sound content in the section and the acoustic feature amount representing the acoustic feature of the sound content in the section after the switching timing, the acoustic feature of the sound content in the section before and after the switching timing. A change amount calculating means for calculating a change amount of the amount; and a changing means for performing processing according to the change amount calculated by the change amount calculating means on the sound content in the section before and after the switching timing. There is provided a signal processing device characterized by comprising:

  If the sound content that has been processed by the signal processing device of the present invention is played back, the sound feature of the preceding sound content to the sound feature of the subsequent sound content are processed according to the processing content by the changing means, with reference to the switching timing of the sound content. In this manner, the acoustic feature changes with time, and thereby the audibility in the section before and after the switching timing can be controlled. For example, if the change amount is calculated by the change amount calculation means so as to smoothly change in time from the acoustic feature of the preceding sound content to the acoustic feature of the subsequent sound content, the occurrence of an auditory gap or the like can be avoided. it can. In addition, since the analysis target by the analysis unit is limited to the sound content in the section before and after the switching timing, it is not necessary to perform a large-scale process like the technique disclosed in Patent Document 1. Furthermore, the processing target by the changing means is limited to the sound content in the section before and after the switching timing. For this reason, even if the sound content in both the section before and after the switching timing is to be processed, the original acoustic features of the sound contents are not impaired over the entire sound content. Absent. In other words, according to the signal processing device of the present invention, an audible impression at the switching timing of these sound contents can be made natural without deteriorating the acoustic characteristics of a plurality of sound contents that are continuously played back. It becomes possible to control.

  As described above, specific examples of acoustic features include reverberation characteristics and volume, and multiple types of acoustic features may be processed. In a more preferred aspect, the signal processing apparatus of the present invention has a specifying means for specifying the processing mode of the changing means, and the change amount calculating means determines the change amount according to the processing mode specified by the specifying means. It is characterized by calculating. According to such an aspect, it becomes possible for the user of the signal processing device to freely control the audibility before and after switching of the sound content.

  In another preferred aspect, the analysis means analyzes acoustic characteristics in a predetermined frequency band, and the change amount calculation means calculates the change amount for the frequency band. According to such an aspect, when an auditory gap or the like is caused by a difference in acoustic characteristics of a specific frequency band, by changing only the acoustic characteristics of that frequency band, It is possible to avoid the occurrence of a hearing gap or the like at the switching timing of the sound content without impairing the characteristics.

  In a further preferred aspect, each of the plurality of sound contents is provided with a flag indicating whether or not the change of the acoustic feature is permitted, and when the flag is a value allowing the change, the analysis means The change amount calculation means and control means for operating the change means are further included. According to such an aspect, while protecting the sound content whose sound feature is not permitted to be changed by the distribution source of the sound content, the sound feature before and after the content switching timing is set only for the sound content without such restriction. Can be controlled.

  As another aspect for solving the above-described problem, an aspect in which a program that causes a computer such as a CPU (Central Processing Unit) to function as the analysis means, the change amount calculation means, and the change means is conceivable. This is because by operating the computer according to such a program, the computer can function as the signal processing device. As a specific form of providing such a program, the program is stored in a computer-readable recording medium such as a compact disk-read only memory (CD-ROM), a DVD (registered trademark: Digital Versatile Disc), or a flash ROM. There may be a mode of distributing by writing and a mode of distributing by downloading via a telecommunication line such as the Internet.

1 is a block diagram illustrating a configuration of a signal processing device 10 according to an embodiment of the present invention. It is a flowchart which shows the flow of the process which the control part 100 of the same signal processing apparatus 10 performs according to the signal processing program 124a. It is a functional block diagram for demonstrating the function implement | achieved by operating the control part 100 according to the signal processing program 124a. It is a figure which shows the operation example in case the acoustic feature of a control object is a volume. It is a figure which shows the operation example in case the acoustic feature of control object is the length of reverberation time. It is a figure for demonstrating the modification of this invention.

Embodiments of the present invention will be described below with reference to the drawings.
(A: Configuration)
FIG. 1 is a diagram illustrating a configuration example of a signal processing device 10 according to an embodiment of the present invention.
When the signal processing apparatus 10 shown in FIG. 1 continuously plays back each of a plurality of audio data (sampling data string indicating the sound waveform of the music content) each representing the music content, the switching timing of the music content is shown. This is a device that performs signal processing for changing acoustic characteristics so that an audible gap or the like does not occur before and after. Playing a plurality of music contents continuously includes a mode in which a segment section is provided in addition to a mode in which music contents are continuously played without a segment section such as a silent section. As shown in FIG. 1, the signal processing device 10 includes a control unit 100, an external device interface unit 110, a storage unit 120, and a bus 130 that mediates data exchange between these components.

  The control unit 100 is, for example, a CPU. The control unit 100 functions as a control center of the signal processing device 10 by executing the signal processing program 124a stored in the storage unit 120 (more precisely, the nonvolatile storage unit 124). Details of processing executed by the control unit 100 in accordance with the signal processing program 124a will be made clear later to avoid duplication.

  The external device interface unit 110 is a collection of various interfaces (hereinafter referred to as “I / F”) such as a USB (Universal Serial Bus) interface. The external device I / F unit 110 connects various external devices, and exchanges data with the external devices. The external device I / F unit 110 provides the data acquired from the connection destination external device to the control unit 100, and outputs the data provided from the control unit 100 to the connection destination external device. Examples of external devices connected to the external device I / F unit 110 include a storage device such as a USB memory and a sound system.

  In the present embodiment, audio data to be subjected to signal processing (that is, audio data corresponding to each of a plurality of music contents to be continuously played back) is input to the signal processing device 10 via the external device I / F unit 110. The For example, when a USB memory storing audio data corresponding to each of a plurality of music contents to be continuously played back and schedule data indicating the playback order of each audio data is connected to the external device I / F unit 110, the external device I The / F unit 110 reads the schedule data and each audio data from the USB memory, and gives them to the control unit 100. Thereafter, the control unit 100 once writes the schedule data and each audio data into the nonvolatile storage unit 124, reads out the audio data from the nonvolatile storage unit 124 in the order indicated by the schedule data, performs the signal processing, and uses the processed audio data. Overwrite. The processed audio data stored in the non-volatile storage unit 124 in this way is the non-volatile storage unit in the order indicated by the schedule data in response to a user giving a reproduction start instruction via an operation unit (not shown). The data is read from 124, output to a sound system connected to the external device I / F unit 110, and reproduced as sound. In the present embodiment, a case where a plurality of audio data and schedule data are separate data will be described. However, it is a matter of course that the data may be a single piece of data.

  As shown in FIG. 1, the storage unit 120 includes a volatile storage unit 122 and a nonvolatile storage unit 124. The volatile storage unit 122 is a volatile memory such as a RAM (Random Access Memory). The volatile storage unit 122 is used by the control unit 100 as a work area when the signal processing program 124a is executed. The non-volatile storage unit 124 is a non-volatile memory such as a flash ROM. The nonvolatile storage unit 124 stores in advance a signal processing program 124a that causes the control unit 100 to execute a process that clearly shows the characteristics of the present invention. In the present embodiment, the control unit 100 reads the signal processing program 124 a from the nonvolatile storage unit 124 to the volatile storage unit 122 when the power (not shown) of the signal processing apparatus 10 is turned on, and starts executing the signal processing program 124 a. The control unit 100 operating in accordance with the signal processing program 124a reads out audio data in the order indicated by the schedule data in response to a processing start instruction given by the user via an operation unit (not shown). The signal processing shown is started.

  FIG. 2 is a flowchart showing the flow of signal processing executed by the control unit 100 in accordance with the signal processing program 124a. As shown in FIG. 2, the control unit 100 operating according to the signal processing program 124a sequentially reads the audio data to be processed until the end is detected (that is, until the determination result in Step SA100 is Yes) If the determination result in step SA100 is Yes, the control unit 100 refers to the schedule data and determines whether there is subsequent audio data (step SA110). When the determination result in step SA110 is “Yes” (that is, when there is subsequent music content), the control unit 110 detects the end time of the audio data to be processed as the switching timing of the music content. Then, the processes after step SA120 are executed. On the other hand, when the determination result of step SA110 is “No”, the signal processing is terminated without executing the processing of step SA120.

  In step SA120, the control unit 100 analyzes audio data for a predetermined time at the end of the music content (hereinafter referred to as preceding music content) to be played before the switching timing and analyzes a section corresponding to the predetermined time ( Hereinafter, for the preceding music content analysis section), an acoustic feature amount representing an acoustic feature predetermined as a control target is calculated. The time length of the analysis section may be set to a suitable value by performing experiments as appropriate. Moreover, since the analysis method for calculating the acoustic feature amount differs depending on the type of the acoustic feature to be controlled, the details will be clarified in the operation example.

  In step SA130 subsequent to step SA120, the control unit 100 analyzes the audio data for a predetermined time at the beginning of the music content (hereinafter referred to as the subsequent music content) to be reproduced following the switching timing, and performs the predetermined processing. An acoustic feature amount representing an acoustic feature to be controlled in a time section corresponding to time (hereinafter, an analysis section of subsequent music content) is calculated. In this embodiment, calculation of the acoustic feature amount for the analysis section of the subsequent music content (step SA120) is performed after calculation of the acoustic feature amount for the analysis section of the preceding music content (step SA120). The execution order of step SA120 and step SA130 may be interchanged, or both may be executed in parallel.

  In step SA140 subsequent to step SA130, the control unit 100 determines, for each of the preceding music content and the subsequent music content, from the acoustic feature in the analysis section before the switching timing to the acoustic feature in the subsequent analysis section. The amount of change of the acoustic feature amount when smoothly changing the time is calculated for each time. Next, the control unit 100 processes the audio data of each analysis section according to the change amount calculated in step SA140, and writes the processed audio data in the nonvolatile storage unit 124 (step SA150). In addition, since various aspects can be considered also about the calculation aspect of change amount in step SA140, and the process aspect in step SA150 according to the kind of acoustic feature to be controlled, details are clarified in an operation example. In step SA160 subsequent to step SA150, the control unit 100 sets the audio data to be processed as the audio data of the next music content indicated by the schedule data, and executes the processes after step SA100 again.

As described above, the control unit 100 operating according to the signal processing program 124a includes the analysis unit 124a1 that executes the processing of steps SA120 and SA130, the change amount calculation unit 124a2 that executes the processing of step SA140, and the processing of step SA150. It functions as changing means 124a3 for executing processing (see FIG. 3). In the present embodiment, each unit shown in FIG. 3 is realized by a software module. However, each unit may be realized by a hardware module such as an electronic circuit.
The above is the configuration of the signal processing apparatus 10.

(B: Operation)
Next, the operation of the signal processing apparatus 10 will be described by taking as an example the case where the acoustic feature to be controlled is volume and the case where the acoustic feature is reverberation characteristics (more specifically, the length of reverberation time).
(B-1: Operation when the acoustic feature to be controlled is volume)
First, as shown in FIG. 4A, the preceding music content is the content A and the subsequent music content is the content B. As shown in FIG. 4B, the volume VA in the analysis section of the content A is better. The operation of the signal processing apparatus 10 will be described by taking as an example a case where the volume is higher than the volume VB in the analysis section of the content B (that is, VA> VB). Although FIG. 4A illustrates a case where no section such as a silent section is provided between the content A and the content B, it is needless to say that a section may be provided.

  As described above, when the control unit 100 detects the switching timing from the content A to the content B (determination result in step SA100: Yes and determination result in step SA110: Yes), the control unit 100 executes the processes after step SA120. . In step SA120, the control unit 100 calculates the volume VA in the analysis section of the content A, and in step SA130 subsequent to step SA120, calculates the volume VB in the analysis section of the content B. As a method for calculating the volume in each analysis section, a known method such as calculating the acoustic energy in each section (for example, the arithmetic mean of the square values of the sample data) and using the acoustic energy as the volume is appropriately used. good.

  In step SA140 subsequent to step SA130, the control unit 100 changes the volume change amount at each time in the analysis section of the content A so that the volume changes smoothly with the timing of switching between the content A and the content B. The change amount of the volume at each time in the analysis section of the content B is calculated. In the present embodiment, the control unit 100 determines the volume at the start point and the end point in two-dimensional coordinates from the start point of the analysis section of the content A to the end point of the analysis section of the content B with time on the horizontal axis and volume on the vertical axis. The amount of change in volume at each time is calculated so that the volume changes along a time change curve that passes through the volume (in the example shown in FIG. 4B, a straight line indicated by a one-dot chain line).

  Specifically, for each time (sampling timing) in each analysis section, the control unit 100 calculates the difference between the square root of the value indicated by the time change curve at that time and the value of the sampling data at the time as the change amount. To do. This is because the amplitude of the audio data can be changed by adding the change amount to the sample data at each time in the subsequent steps SA140 and SA150. In this operation example, a straight line is used as the time change curve, but any curve may be used as long as it is a smooth curve. Specifically, Hermite has three points: the volume at the start point of the analysis section of the preceding music content, the volume at the end point of the analysis section of the subsequent music section, and the volume located between the two volumes at the boundary of both sections A curve obtained by interpolation or spline interpolation can be considered.

  In step SA150 subsequent to step SA140, control unit 100 changes the amplitude of the audio data of content A so that the volume in the analysis section of content A changes over time according to the change amount calculated in step SA140. A process (a process of adding the change amount corresponding to the time to the sample data at each time) is written into the nonvolatile storage unit 124, and the volume in the analysis section of the content B is set to the change amount calculated in step SA140. Therefore, a process of changing the amplitude of the audio data of the content B so as to change with time is performed and written in the nonvolatile storage unit 124. When the audio data of the content A and the audio data of the content B processed in the manner described above are continuously played back, the volume is changed from VA to VB at one point in FIG. The time changes smoothly according to the straight line shown by the chain line, and the difference in volume between the two contents is alleviated. For this reason, an audible gap or the like due to a sudden change in volume before and after the switching timing does not occur.

(B-2: Operation when the acoustic feature to be controlled is a reverberation characteristic)
Next, an operation when the acoustic feature to be controlled is a reverberation characteristic will be described.
When playing back multiple music contents in succession, if there is a difference between the amount of reflected sound in the preceding music content and the amount of reflected sound in the subsequent music content, the atmosphere of the sound field becomes large before and after the switching timing of the music content. It changes and a hearing gap occurs. As shown in FIG. 5A, the reflected sound includes an initial reflected sound and a reverberant sound. The initial reflected sound refers to a sound that reaches the listener through the first reflection by a wall or the like after being emitted from the sound source, and the reverberant sound refers to a sound that reaches the listener through a plurality of reflections. Reverberation is also called late reflection. When the acoustic feature to be controlled is a reverberation characteristic, the amount of change in the reverberation sound of the music content is calculated so that the amount of the reverberation sound of the music content to be continuously played is naturally connected, and the reverberation sound is removed or added. What is necessary is just to make it make the signal processing apparatus 10 perform.

  More specifically, when the acoustic feature to be controlled is a reverberation characteristic, the audio data to be processed (the audio data for a predetermined time at the end of the preceding music content in step SA120, in step SA120 and step SA130 described above). In step SA130, the control unit 100 may be configured to analyze the audio data for a predetermined time from the beginning of the subsequent music content and calculate the reverberation time for each. The reverberation time is one of the indexes for evaluating the reverberation characteristics, and the longer the reverberation time, the greater the amount of reverberation sound. A known technique may be adopted as appropriate for the calculation method of the reverberation time. Specifically, a method for calculating the reverberation time by estimating the power of the initial reflected sound and the reverberation sound from the audio data to be processed (see Non-Patent Document 1), analyzing the audio data to be processed, and analyzing the signal energy. A method for calculating the reverberation time from the attenuation rate in the attenuation unit (see Non-Patent Document 2), a method for calculating the reverberation time by maximum likelihood estimation for the attenuation unit (Non-Patent Document 3), etc. Can be considered. Moreover, you may employ | adopt the method (Nonpatent literature 4) which estimates the acoustic energy of reverberation sound instead of the reverberation time, and estimates the amount of reverberation sound directly.

  The control unit 100 changes the amount of reverberation sound at each time in the analysis section of the content A and each change in the analysis section of the content B so that the reverberation characteristics change smoothly between the switching timings of the content A and the content B. The amount of change of the reverberant sound at the time is calculated (step SA140). As shown in FIG. 5B, the reverberation time (the reverberation time of the preceding music content (content A)) TA calculated in step SA110 is greater than the reverberation time TB of the subsequent music content (content B). It is assumed that the content is longer (that is, the content A has more reverberant sounds). In this case, the control unit 100 determines the reverberation time TA at the start point in two-dimensional coordinates from the start point of the analysis interval of the content A to the end point of the analysis interval of the content B, with the horizontal axis representing time and the vertical axis representing the reverberation time. The amount of change of the reverberant sound at each time (in content A) so that the reverberation time changes along the time change curve passing through the reverberation time TB at the end point (in the example shown in FIG. 5B, a straight line indicated by a dashed line). On the other hand, the amount of reverberant sound to be removed and the amount of reverberant sound to be added to content B) are calculated. In the present embodiment, for the content A, the control unit 100 compares the value indicated by the time change curve at the time and the value of the reverberation time calculated at step SA120 for the time in the analysis section (the former is excluded from the latter). For the content B, for each time in the analysis section, the ratio between the value indicated by the time change curve at that time and the value of the reverberation time calculated in step SA130 The amount of change in the reverberant sound at the time.

  In step SA150, control unit 100 executes a process of removing or adding reverberant sound according to the change amount calculated in step SA140. Spectral subtraction is an example of a method for removing reverberant sound. Spectral subtraction is a subtraction process in the frequency domain, and is realized as follows. First, the control unit 100 performs FFT on the audio data to be processed and converts it to frequency domain data. Next, the control unit 100 tracks the time change of the amplitude level for each frequency bin in the FFT, determines that a reverberation sound is added to the frequency bin whose attenuation width is less than a predetermined threshold, and multiplies the change amount. This suppresses the amplitude. This is because, in general, the sound with the reflected sound is attenuated more slowly than the sound without the reflected sound. It should be noted that the amount of reverberant sound removal may be adjusted more finely by adjusting the threshold value or adjusting the amplitude suppression amount (for example, multiplying a constant in addition to the change amount).

  On the other hand, a method for giving a reverberant sound is as follows. First, the control unit 100 generates reverberation sound data by convolving the audio data to be processed with an impulse response having an amplitude corresponding to the change amount and the acoustic energy of the audio data. The reverberation sound data is data representing the “reverberation sound” in FIG. Next, the control unit 100 adds the reverberant sound data generated in the above manner to the audio data to be processed. Note that the amount of reverberant sound added may be adjusted more finely by adjusting the mixing ratio when adding reverberant sound data to the audio data to be processed or adjusting the length of the impulse response.

  In this operation example, processing for removing the reverberation sound is performed for the analysis section of the content A while increasing the removal amount gradually, and processing for adding the reverberation sound for the analysis section of the content B while gradually decreasing the addition amount is performed. Done. When the audio data of the content A and the audio data of the content B are continuously played back, the reverberation time is smooth from TA to TB according to the straight line shown by the alternate long and short dash line in FIG. It changes over time and the difference in the reverberation characteristics of both contents is alleviated. For this reason, an audible gap or the like due to a sudden change in the reverberation characteristic does not occur before and after the switching timing.

  As described above, according to the signal processing device 10 of the present embodiment, it is possible to smoothly change the acoustic features such as the volume from the preceding music content to the subsequent music content with time, based on the switching timing of the music content. it can. For this reason, when the music content is switched, an audible gap or the like due to a difference in acoustic characteristics between the two contents is avoided. In addition, since the analysis target by the signal processing apparatus 10 is limited to audio data in the analysis section before and after the switching timing of the music content, the audio of each of the preceding and subsequent music contents as in the technique disclosed in Patent Document 1 is used. There is no need to perform large-scale processing on the entire data. Since the processing target by the signal processing apparatus 10 is also limited to the audio data in the analysis section before and after the switching timing of the music content, the original acoustic features of the preceding and succeeding music contents are impaired throughout the entire music content. It will never happen. In other words, according to the signal processing device 10 of the present embodiment, an audible impression at the switching timing of the content is naturally obtained without impairing the acoustic characteristics of each of the plurality of music contents that are continuously played back. (That is, generation of an auditory gap or the like can be avoided).

(C: deformation)
Although one embodiment of the present invention has been described above, it goes without saying that the following modifications may be added to this embodiment.
(1) In the above embodiment, the case has been described where the plurality of sound contents to be continuously played back are music contents. However, the plurality of sound contents to be played back continuously is not limited to music contents, and novels and guidance sentences. The sound may be read aloud, such as environmental sounds such as forest sounds and wave sounds, and disturbing sounds (the contents are rendered meaningless by dividing the sound into multiple frames and rearranging the frames, etc.) Or a masker sound such as In the above-described embodiment, a case has been described in which processing is performed so that the acoustic feature to be controlled changes smoothly for both analysis sections before and after the switching timing of music content. However, the processing may be performed so that the acoustic feature smoothly changes in any one of the analysis section before the music content switching timing and the subsequent analysis section. For example, when the acoustic feature to be controlled is a volume, the acoustic feature may be changed only for the analysis section of the preceding music content as indicated by a dashed line in FIG. 6B, the acoustic feature may be changed only for the analysis section of the subsequent music content as indicated by the alternate long and short dash line in FIG.

(2) In the above embodiment, audio data representing each of a plurality of music contents to be continuously played is input to the signal processing apparatus 10 via the external device I / F unit 110, and the external device I / F unit 110 is A case has been described in which processed audio data is output to a sound system or the like. However, instead of the external device I / F unit 110, a communication I / F unit such as a NIC (Network Interface Card) is provided in the signal processing device 10, and an electric communication line such as the Internet is connected to the communication I / F unit. Then, audio data representing each of a plurality of music contents to be played back may be input to the signal processing device 10 via the telecommunication line. Similarly, processed audio data may also be output via the telecommunication line. According to such an aspect, audio data representing each of a plurality of music contents to be continuously played back is received via the telecommunication line, and an auditory gap or the like is provided at the timing of switching the contents without impairing the acoustic characteristics of each music content. ASP (Application Service) that returns audio data that has been processed to prevent occurrence of
Provider) format signal processing services can be provided.

(3) In the above embodiment, the generation of an auditory gap or the like is avoided by processing the audio data of each analysis section so that the acoustic features smoothly change in time in the analysis section before and after the switching timing of the music content. Explained the case. However, the audio data of each analysis section may be subjected to processing for changing the acoustic feature so that the switching timing of the music content is emphasized, and the difference of the acoustic feature in the analysis section before and after the switching timing is emphasized. As described above, the audio data of each analysis section may be subjected to processing for changing acoustic characteristics. In this case, the auditory sensation gap or the like is emphasized, but it is possible to perform an effect using this.

  A specific example of the aspect in which the acoustic feature is changed to the audio data of each analysis section so that the switching timing of the music content is emphasized is as follows. When multiple music contents that are played back continuously are BGM (Back Ground Music) that flows in each scene of a movie or video game, the switching timing is emphasized and the listener is strongly reminded of the switching of the scene. be able to. For example, when a BGM flowing in an open space such as a field is reproduced following a BGM flowing in a narrow cave, as shown in FIG. 6 (c), there is a difference in reverberation time between the preceding BGM and the succeeding BGM. If there is no sound, there will be no change in the sense of sound spread, and it will be difficult to recall the scene change from BGM alone (the vast field spread after leaving the cave). On the other hand, as shown by the alternate long and short dash line in FIG. 6C, the reverberation time is abruptly increased to emphasize the sense of sound spread, and then gradually brought closer to the original value to emphasize the BGM switching timing. In addition, the listener can be reminded of the switching of the scene. Note that, as shown by a two-dot chain line in FIG. 6C, the same effect can be obtained by reducing the reverberation time rapidly and then gradually bringing it closer to the original value.

  The specific example of the aspect which emphasizes the difference of the acoustic feature in the analysis section before and after the switching timing is as follows. For example, when the sound content B is reproduced after the sound content A, the sound feature to be controlled is a volume, and the volume VA of the sound content A is larger than the volume VB of the sound content B, FIG. As shown in (d), after the volume is temporarily increased to a larger value (VMAX) in the analysis section for the sound content A, the volume is rapidly decreased to a value (VMIN) smaller than the volume VB at the content switching timing. Thereafter, a mode in which the volume is gradually increased to the volume VB in the analysis section for the sound content B can be considered. As shown in FIG. 6 (d), the volume is not changed discontinuously at the switching timing, but as shown in FIG. 6 (e), the volume VA → volume VMAX → The volume may be continuously changed in the order of the intermediate value between the volume VA and the volume VB → the volume VMIN → the volume VB. According to these aspects, the difference in acoustic characteristics between the two contents is emphasized before and after the content switching timing, and it is possible to perform an effect using this.

  As described above, according to the present invention, it is possible to perform an effect that has not been achieved in the past by controlling the acoustic features in the analysis section before and after the switching timing of the sound content. It should be noted that even in the aspect of performing such an effect, there is no change in the analysis by the signal processing device 10 and the processing of changing the acoustic feature is limited to the sound content in the analysis section before and after the switching timing. The original acoustic features of each subsequent sound content are not impaired throughout the sound content.

(4) The signal processing apparatus 10 is provided with a specifying unit that specifies the processing mode of the changing unit 124a3, and the change amount calculating unit 124a2 calculates the change amount according to the processing mode specified by the specifying unit, The changing unit 124a3 may be caused to perform the processing of the mode specified by the specifying unit on the sound content. Here, regarding the mode of processing executed by the changing unit 124a3, (a) whether to analyze both analysis sections before and after the switching timing, or one of the analysis section before and after the switching timing. Classification of the analysis section to be processed, and (b) processing to alleviate differences in acoustic features, or emphasize differences in acoustic features (or switching timing). It is possible to classify the contents of processing, such as whether it is processing. The designation means may designate an analysis section to be processed, or may designate processing contents. Further, both the analysis section to be processed and the processing content may be specified. As a specific example of such a designation means, a user interface unit including an input device such as a touch panel and a mouse keyboard and a display device can be considered.

(5) The analysis unit 124a1 analyzes the acoustic characteristics of the audio data to be analyzed in a predetermined frequency band, the change amount calculation unit 124a2 calculates the change amount in the frequency band, and the change unit 124a3 The audio data may be updated so that the acoustic feature amount of the band changes by an amount corresponding to the change amount. For example, when a plurality of sound contents to be continuously played is centered on voice such as vocal music, and the acoustic feature to be controlled is a reverberation characteristic, the acoustic feature is not obtained for the voice band (for example, 125 Hz to 2 kHz). The acoustic feature is controlled only for the higher sound range than the voice band without changing. Since the sound reverberation (spreading) is more easily detected in the high sound range, according to such an aspect, the auditory gap or the like is alleviated without greatly changing the sound quality (auditory impression) of the entire sound content. It becomes possible.

  Further, the analysis unit 124a1 performs band division to divide into a plurality of predetermined band components on the audio data to be analyzed to specify acoustic features for each band, and the change amount calculation unit 124a2 The change amount may be calculated every time, and the changing unit 124a3 may be caused to execute processing for changing the acoustic feature amount by an amount corresponding to the change amount for each band. In this case, although the common acoustic feature is changed in each band, the processing mode of the process of changing the acoustic feature may be changed for each band, or the different acoustic feature may be changed for each band.

Although specific acoustic features are changed in each band, a specific example of an aspect in which the processing aspect of changing the acoustic feature is different for each band is as follows. The aspect which changes first, changes a low-pitched range later, ie, the aspect which changes the timing to change with a high-pitched-range (for example, 2 kHz-4 kHz) and a low-pitched-range (band | band lower than 2 kHz) is mentioned. As described above, since the reverberation is easier to detect in the high sound range, changing the high sound range first can avoid the sudden change in sound quality and reduce the auditory gap. A specific example of how to change the acoustic features that differ for each band is when multiple sound contents to be played back are mainly voices such as vocal songs, and you want to control both the reverberation and the volume of the vocals. In addition, with respect to the voice band, a mode in which the volume is set as the acoustic feature to be controlled and a reverberation characteristic is set as the acoustic feature to be controlled in the high pitch range can be considered.
According to such an aspect, it becomes possible to finely control the audibility before and after switching of the sound content for each band. Of course, this modification may be combined with the modification (1) or (3) described above.

(6) In the above-described embodiment, a case has been described in which a process for unconditionally changing an acoustic feature is performed on each of a plurality of audio data input to the signal processing device 10. Some music content, such as music clips, has been adjusted for the commercial image strategy to be unique to singers and performers. Changing features can cause problems. This is because if the sound characteristics are changed through a video site or the like, the image strategy may be hindered. Therefore, for music content for which it is not desirable to change the acoustic feature, the audio data is distributed to the distribution source with a flag set to indicate that the change of the acoustic feature is not permitted, while the control unit 100 Are functioned as the analysis unit 124a1, the change amount calculation unit 124a2, and the change unit 124a3, and the value of the flag given to the audio data to be processed is a value indicating that the change of the acoustic feature is permitted. In addition, the non-volatile storage unit 124 may store a signal processing program that functions as a control unit that operates the analysis unit 124a1, the change amount calculation unit 124a2, and the change unit 124a3.

  DESCRIPTION OF SYMBOLS 10 ... Signal processing apparatus, 100 ... Control part, 110 ... External apparatus I / F part, 120 ... Memory | storage part, 122 ... Volatile memory | storage part, 124 ... Nonvolatile memory | storage part, 124a ... Signal processing program, 130 ... Bus.

Claims (5)

Analysis means for analyzing the acoustic characteristics of each sound content in a section before and after the switching timing of the plurality of sound content to be played back continuously;
Based on the acoustic feature amount representing the acoustic feature of the sound content in the section before the switching timing and the acoustic feature amount representing the acoustic feature of the sound content in the section after the switching timing, before and after the switching timing. A change amount calculating means for calculating a change amount of the acoustic feature amount of the sound content in the section;
Change means for performing processing according to the change amount calculated by the change amount calculation means for the sound content in the section before and after the switching timing;
A signal processing apparatus comprising:   The signal processing apparatus according to claim 1, wherein the acoustic feature is a reverberation characteristic.   The signal processing apparatus according to claim 1, wherein the acoustic feature is a volume. Having a specifying means for specifying the processing mode of the changing means;
The signal processing apparatus according to claim 1, wherein the change amount calculation unit calculates the change amount according to a processing mode designated by the designation unit. Each of the plurality of sound contents is provided with a flag indicating whether or not the change of the acoustic feature is permitted,
5. The apparatus according to claim 1, further comprising a control unit that activates the analysis unit, the change amount calculation unit, and the change unit when the flag is a value that permits the change. A signal processing device according to 1.

Images