JP5034976B2 - Audio playback device and audio playback control program - Google Patents

Description

  The present invention relates to an audio playback device and an audio playback control program.

  In video games, etc., various sounds such as music, sound effects, speech, etc. are used. The tempo (speed) and pitch (pitch) are controlled according to the situation, or the tempo and pitch can be arbitrarily set by the user (player). There is a need to make it operate.

Conventionally, in order to individually and arbitrarily control the tempo and pitch during playback of an audio file, complex audio data processing is required, which can be realized by a personal computer having high processing capability, but is limited to video game devices and the like. It has been difficult to achieve with the resourced devices. In this regard, for music and sound effects based on the MIDI (Musical Instruments Digital Interface) standard, the tempo and pitch can be easily controlled simply by changing the data description. It was. Note that since the MIDI standard controls the pitch, time, volume, and the like for reproducing a sound source in time series, a general-purpose audio file such as a digital recording of live performances or speech cannot be reproduced.
Japanese Patent Laid-Open No. 6-59679

  As described above, it is difficult to control the tempo and pitch individually and arbitrarily for general-purpose audio files such as digital recordings of live performances and speeches in devices with limited resources such as game devices. There is a demand for a mechanism that can be realized by simple processing.

  On the other hand, a new sound source creation method called “granular synthesis” has been proposed (see, for example, Patent Document 1), but it is limited to sound source creation, and is reproduced accurately in time series such as music and speech. It was not applied to playback of audio files that were required to be performed.

  The present invention has been proposed in view of the above-described conventional problems, and the object of the present invention is to individually set the tempo and pitch for general-purpose audio files such as music, sound effects, and speeches with a simple configuration. Another object of the present invention is to provide an audio reproduction device and an audio reproduction control program that can be arbitrarily controlled. Another object is to freely control the tempo and pitch by the movement of the controller.

In order to solve the above problems, according to the present invention, as described in claim 1, an extraction start position setting unit that sets an extraction start position for extracting grains from an audio file, and a grain length of the grains to be extracted are set. A grain length setting unit to be set, a grain extraction unit for extracting the grain of the grain length from the extraction start position on the audio file, a reproduction unit for reproducing the extracted grain according to a set pitch, and when the reproduction of the grain is completed An extraction start position update unit that gives a reference position on the audio file according to a set tempo and instructs the extraction start position setting unit to update the extraction start position , and the extraction start position setting unit includes: Obtain a randomized value based on the current tempo and / or pitch from the control table and The starting position is taken out based on the randomized value, and the gist of the audio reproduction device which makes random manner back and forth relative to a reference position on the audio file.

In addition, as described in claim 2 , the audio playback device according to claim 1 , further comprising: a grain length update unit that instructs the grain length setting unit to update the grain length when the grain reproduction is completed. The length setting unit can acquire the grain length based on the current tempo and / or pitch from the control table.

In addition, as described in claim 3 , in the audio reproduction device according to claim 1 or 2 , the grain extraction unit includes a plurality of different extraction start positions based on a predetermined overlap phase. The grain can be extracted and the reproduction unit can synthesize each grain after multiplying each grain by a predetermined window function.

According to a fourth aspect of the present invention, there is provided the audio reproduction device according to the third aspect , further comprising a window function zero value determination unit that detects the completion of grain reproduction when the window function becomes zero. be able to.

According to a fifth aspect of the present invention, there is provided a computer that constitutes the device, a take-out start position setting means for setting a take-out start position for taking out the grains from the audio file, a grain length setting means for setting the grain length of the grains to be taken out, Grain extraction means for extracting the grain having the grain length from the extraction start position on the audio file, reproduction means for reproducing the extracted grain according to a set pitch, and an audio file corresponding to the set tempo when the reproduction of the grain is completed An upper reference position is given to function as an extraction start position update means for instructing an update of the extraction start position , and the extraction start position setting means obtains a randomized value based on the current tempo and / or pitch from the control table And based on the randomization value Extraction start position Te, and can be configured as a random number to the voice reproduction control program causes back and forth relative to a reference position on the audio file.

  In the audio reproduction device and the audio reproduction control program of the present invention, the tempo and pitch can be individually and arbitrarily controlled for general-purpose audio files such as music, sound effects and speeches with a simple configuration. In addition, the tempo and pitch can be freely controlled by the movement of the controller.

  Hereinafter, preferred embodiments of the present invention will be described. Although an example applied to a game device will be described, it goes without saying that the present invention can be applied to other devices.

  FIG. 1 is a diagram showing a configuration example of a game device according to an embodiment of the present invention.

  In FIG. 1, the game apparatus 1 includes a wireless communication unit 101, and communicates with the wireless communication unit 23 of the controller 2 that the player holds and operates, so that the operation switch 21 and the three-dimensional acceleration sensor 22 of the controller 2 are in a state. Can be acquired as data.

  On the other hand, the game apparatus 1 has a controller operation determination unit 102 that determines an operation content based on an input from the controller 2 received by the wireless communication unit 101, and a game progress based on the operation content determined by the controller operation determination unit 102. And a granular synthesis processing unit 105 that outputs sound based on the principle of granular synthesis under the control of the game control unit 104.

  The controller operation determination unit 102 manages tempo and pitch data for audio reproduction. The tempo and pitch are set by the game control unit 104 during audio reproduction, and the player's operation switch 21 of the controller 2 controls the player. It is set by a direct instruction, and is further set by the acceleration change interpretation unit 103 interpreting operations such as swinging the controller 2 up and down, left and right, back and forth, rotating, and throwing while rotating. Note that the tempo is the audio playback speed. When the tempo is “1”, the tempo is played at the same speed as the original voice. For example, when the tempo is “2”, the tempo is played at twice the original speed ( As a result, the entire playback time is half), and when the tempo is “0.5”, playback is performed at half the speed of the original sound (as a result, the total playback time is doubled). The pitch is the pitch of the reproduced voice. When the pitch is “1”, the pitch is reproduced with the same pitch as the original voice. For example, when the pitch is “2”, the pitch is reproduced with twice the pitch of the original voice. When the pitch is “0.5”, the sound is reproduced with half the pitch of the original sound.

  The granular synthesis processing unit 105 inputs a digital data state audio file held in the memory 115 and outputs a digital data state audio signal. The audio file on the memory 115 is stored in the memory 115 together with the control program and various data from the medium 116 such as a CD (Compact Disc) and a DVD (Digital Versatile Disk) by the media reading unit 117, and the microphone 118, A / Digital recording may be performed via the D (Analog / Digital) conversion unit 119 and the effector unit 120. The effector unit 120 provides sound range adjustment and special effects.

  In addition, an audio output signal of digital data from the granular synthesis processing unit 105 is sequentially converted into sound by the speaker 125 via the panning unit 121, the effector unit 122, the volume changing unit 123, and the D / A (Digital / Analog) conversion unit 124. Is output. The panning unit 121 distributes the sound to the left and right (L: Left, R: Right) channels of the stereo, and the volume changing unit 123 changes the volume level (volume).

  As an internal configuration of the granular synthesis processing unit 105, an instruction to update a grain extraction start position from an audio file on the memory 115 is given at a timing when a window function 0 value determination unit 113 described later determines a 0 value of the window function. An extraction start position update unit 106, an extraction start position setting unit 107 that sets a grain extraction start position in accordance with an update instruction from the extraction start position update unit 106, and a grain length that sets a grain length that is the length of the grain to be extracted A setting unit 109 and an overlap phase setting unit 110 that sets a phase for changing a phase and extracting a plurality of grains for overlap (details will be described later) are provided. Here, the grain is minute audio data (several tens to several hundreds [ms]) extracted from the audio file. The extraction start position update unit 106 manages a locator value corresponding to the recording time of the audio file, and this locator value changes at a speed according to the tempo given from the controller operation determination unit 102 (details) Will be described later). Further, the extraction start position setting unit 107 manages a randomized value for randomly moving the grain extraction start position back and forth. The randomized value and the grain length set by the grain length setting unit 109 are: It is set according to the current tempo and pitch by the control table 108 (details will be described later).

  Further, the granular synthesis processing unit 105 uses the grain length set by the grain length setting unit 109 from the extraction start position set by the extraction start position setting unit 107 and the phase set by the overlap phase setting unit 110. A grain extraction unit 111 that extracts a plurality of grains from an audio file, and a window that multiplies a predetermined window function for each grain while reproducing each of the extracted grains at a speed based on a pitch given from the controller operation determination unit 102. A playback / window function processing unit 112 that performs function processing and synthesis, a window function 0 value determination unit 113 that detects a timing at which the window function in the playback / window function processing unit 112 becomes 0 at the rear end, and a window function 0 The grain length is set to update the grain length at the timing when the value judging unit 113 judges the zero value. And a grain length update section 114 which instructs the 109. Details of the window function will be described later.

  Note that each functional unit of the granular synthesis processing unit 105 is realized by a control program on the memory 115, and data of the process is held as various data on the memory 115.

  FIG. 2 is a diagram illustrating a configuration example of the acceleration change interpretation unit 103 of the controller operation determination unit 102. FIG. 2A shows the movement direction of the controller 2. Here, in the global coordinate axes, the horizontal direction is the X direction, the front-rear direction is the Y direction, and the vertical direction is the Z direction.

  FIG. 2B shows a configuration example of the acceleration change interpretation unit 103, and X direction acceleration, Y direction acceleration, and Z direction acceleration data (acceleration data) are input from the controller 2 input received by the wireless communication unit 101. X-direction acceleration acquisition unit 103a, Y-direction acceleration acquisition unit 103b, Z-direction acceleration acquisition unit 103c, X-direction acceleration acquisition unit 103a, Y-direction acceleration acquisition unit 103b, and Z-direction acceleration acquisition unit 103c acquired Threshold processing units 103d to 103f that perform threshold processing for extracting only those exceeding a predetermined threshold. The threshold processing is performed in order to prevent a malfunction by detecting a small movement or the like of the controller 2 not intended by the player. Here, the X-direction acceleration, the Y-direction acceleration, and the Z-direction acceleration in the acceleration data are local coordinate axes attached to the three-dimensional acceleration sensor 22, and the length of the controller 2 as shown in FIG. When the direction is aligned with the global Y axis and a predetermined inclination is obtained, the global coordinate axis and the local coordinate axis are matched. In the following description, in order to simplify the explanation, it is assumed that the global coordinate axis and the local coordinate axis coincide with each other in the initial state, the initial state is maintained in the vibration described later, and the rotation returns to the initial state after the rotation. Shall. In practice, the global coordinate axis and the local coordinate axis are often different. In this case, one of the coordinate axes may be selected, or coordinate conversion from local to global may be performed.

  The acceleration change interpretation unit 103 includes an output of the threshold processing unit 103d, that is, a vibration cycle detection unit 103g that detects a vibration cycle in the horizontal direction from the acceleration change in the X direction, and an output of the threshold processing unit 103e, that is, an acceleration in the Y direction. A vibration cycle detection unit 103h that detects a vibration cycle in the front-rear direction from the change, and a vibration cycle detection unit 103i that detects a vibration cycle in the vertical direction from the output of the threshold processing unit 103f, that is, an acceleration change in the Z direction, are provided. Further, a rotation cycle detection unit 103j that detects a rotation cycle about the longitudinal direction of the controller 2 from the outputs of the threshold processing unit 103d and the threshold processing unit 103f, that is, the change in the acceleration data in the X direction and the change in the acceleration data in the Z direction. And a rotary throw detection unit 103k that detects a state when throwing while quickly rotating around the longitudinal direction of the controller 2 from the outputs of the threshold processing unit 103d and the threshold processing unit 103f. Further, the acceleration change interpretation unit 103 changes the tempo and / or pitch based on a predetermined operation rule from the outputs of the vibration cycle detection units 103g to 103i, the rotation cycle detection unit 103j, and the rotation throw detection unit 103k. A change unit 103l is provided.

  FIG. 3 is a diagram showing an example of acceleration change. FIG. 3A shows a change in acceleration data in a corresponding direction when the controller 2 is swung in any of the left, right, front, back, and top and bottom directions. The vibration period detectors 103g to 103i of the acceleration change interpreter 103 detect the vibration period in the corresponding direction from the change in the acceleration data, and the tempo / pitch changer 103l is, for example, the vibration in the vertical direction (Z direction). Is associated with the tempo, and if it is faster than the current tempo, the tempo is gradually increased, and if it is slower, the tempo is gradually decreased.

  (B) shows changes in the acceleration data of the X-direction acceleration and the Z-direction acceleration when the controller 2 is rotated counterclockwise about the longitudinal direction (left-rotating around the Y axis). Since the three-dimensional acceleration sensor 22 of the controller 2 detects the gravitational acceleration, a phase difference between the acceleration data waveforms of the X-direction acceleration and the Z-direction acceleration occurs with the rotation, and can detect whether the rotation is right-handed or left-handed. . The tempo / pitch changing unit 103l, for example, causes the right rotation to correspond to the pitch increase and the left rotation to correspond to the pitch decrease, so that when the right rotation is detected, the pitch is gradually increased according to the rotation period, When left rotation is detected, the pitch is gradually lowered according to the rotation period.

  (C) shows the change in the acceleration data of the X-direction acceleration or the Z-direction acceleration when the controller 2 is thrown up while rapidly rotating (rotating around the Y axis) about the longitudinal direction. When the controller 2 is quickly rotated, a centrifugal force larger than the gravitational acceleration is applied to the three-dimensional acceleration sensor 22, and the acceleration at the time τ exceeding the predetermined time T becomes substantially constant beyond the preset acceleration a. Since there is a period, such a rotation throw can be detected. However, the period in which the acceleration exceeds the acceleration a by the centrifugal force and is substantially constant occurs almost simultaneously in the acceleration data of the X-direction acceleration and the Z-direction acceleration. Therefore, the rotation direction cannot be detected from the phase difference of the acceleration data change during this period. . However, since a gentle rotation is accompanied immediately before the rotation throw, this can be set as the rotation direction of the rotation throw. For example, the tempo / pitch changing unit 103l sets a control program in advance so that the right rotation changes rapidly, and the left rotation changes sharply, for example, at a high speed rotation in rotation throwing. The pitch is rapidly increased when a rotational throw is detected, and the pitch is rapidly lowered when a left rotational throw is detected. By doing in this way, an effect like a glissando can be produced by reproducing the sound.

  2B detects the rotation of the controller 2 around the X axis and the rotation of the controller 2 around the Z axis from the difficulty of the operation by the player. Although it does not have a configuration, when such an operation is assumed, it can be similarly detected based on the above-described principle and used for tempo / pitch control.

  Returning to FIG. 1, the extraction start position update unit 106 of the granular synthesis processing unit 105 starts to extract the locator value at that time as the reference of the extraction start position at the time of the window function 0 value determination of the window function 0 value determination unit 113. The position setting unit 107 is instructed to update the take-out start position. The locator value increases at a speed corresponding to the tempo (the locator moves at a speed corresponding to the tempo), and the locator value at the timing when the window function reaches 0 is used as a reference for the new extraction start position. The 4A and 4B are explanatory diagrams of the locator and take-out position update. FIG. 4A shows the case of tempo “1” as an example, and FIG. 4B shows the case of tempo “2” as an example. In the case of tempo “1” in (a), the locator value is the same value as the actual time during processing. In the case of tempo “2” in (b), the locator value is twice the actual processing time. For example, if an audio file for one second (1000 [msec]) is 0 (start position) to 1000 (end position), the actual processing time is 1 [msec] and “1” when the tempo is “1”. If the tempo is “2”, the actual processing time is 1 [msec] and increases by “2”.

  4, in any case, the extraction start position update unit 106 finishes reproducing the time point when the window function at the rear end of the window function period becomes zero (basic grain (first grain in the case of overlap)). The take-out start position setting unit 107 is instructed to update the take-out position. For example, if the grain length being reproduced is 100 [msec] and the pitch is “1”, the window function takes a value of 0 with a period of 100 [msec]. Therefore, when the tempo is “1”, the reference is “100”. When the tempo is “2”, the reference is moved by “200”.

Returning to FIG. 1, the extraction start position setting unit 107 sets the extraction start position at the beginning of the audio file or when the extraction start position update unit 106 is instructed to update the extraction start position. The extraction start position setting unit 107 sets the start position as the extraction start position when processing from the beginning of the audio file, and when instructed to update the extraction start position from the extraction start position update unit 106, based on the randomized value, The position is moved back and forth at random from the reference (locator value) of the extraction start position given from the extraction start position update unit 106. For example, when the randomized value is “0”, the reference extraction start position becomes the extraction start position as it is. When the randomized value is “10”, a position randomly determined within a range of −10 to +10 [msec] with respect to the reference extraction start position is the extraction start position. It has been confirmed that by randomizing the extraction start position, the natural feeling for the listener is increased as compared with the case where the extraction start position is regularly set. The randomization value must be such that the next grain picking start position does not come before the previous grain picking start position.
Grain length x tempo ÷ set so as not to exceed pitch. As a result, the take-out start position moves along with the progress of time without returning, and the sound of music or the like can be reproduced faithfully along the time axis.

  Note that, by obtaining a value corresponding to the current tempo / pitch from the control table 108, the sound quality when the tempo or pitch is changed can be maintained finely. However, if there is no problem in sound quality, a fixed value may be set as the randomization value.

  FIG. 5 is a diagram showing an example of the control table 108, in which randomized values and grain lengths corresponding to the tempo and pitch are set. Note that the randomized value and the grain length may be associated only with either the tempo or the pitch, or the randomized value and the grain length may be individually associated with the tempo or the pitch.

  Returning to FIG. 1, the grain length setting unit 109 sets the grain length at the beginning of the audio file or when the window function 0 value determination unit 113 determines the window function 0 value. By obtaining a value corresponding to the current tempo / pitch from the control table 108, the grain length can be finely maintained when the tempo or pitch is changed. However, if there is no problem in sound quality, a fixed value may be set as the grain length.

  6A and 6B are diagrams showing an example of the control table 108 for the grain length, where FIG. 6A is an example of the data structure, and FIG. 6B is the grain length with respect to the tempo (the intermediate value of the values described in the control table 108 is linear interpolation). Represents. That is, when the tempo is “0”, the grain length is 70 [msec], when the tempo is “± 0.5”, the grain length is 80 [msec], and when the tempo is “± 1” or more, the grain length is 100 [msec]. This is an example. In addition, although the example which matched the grain length only with tempo was demonstrated, you may match | combine a grain length only with a combination with a pitch, or a pitch.

  Returning to FIG. 1, the overlap phase setting unit 110 sets a phase difference according to the number of overlaps. The number of overlaps is usually about 4-16. FIG. 7 is a diagram showing an example of overlap, in which the number of overlaps is four. In this case, a total of four grains are extracted from the leading grain with a phase difference of ¼ of the grain length, and each grain is synthesized after performing window function processing during reproduction.

  The overlap is for removing the adverse effect of the window function processing. That is, FIG. 8 is a diagram showing an example of a window function. As shown in FIG. 8A, since the extracted grain has arbitrary levels at both ends, the level rapidly changes when connected to the preceding and succeeding grains. Cause noise. For this reason, window function processing for multiplying the waveform of the grain by the window function is performed to converge both ends of the grain to 0 level. The window function is used during grain playback and is synchronized to match the grain playback length. In other words, the length (cycle) of the grain length may differ between the time of extraction and the time of reproduction depending on the pitch or the like, but the length (cycle) of the window function always matches the grain length at the time of reproduction. Various types of window functions can be used, examples of which are shown in (b) to (d).

  The window function is used to prevent noise in this way, but if both ends of all grains are at 0 level, the sound will be interrupted and heard at the period of the grains. Therefore, by overlapping each individual grain, it is possible to prevent the sound from being interrupted.

  Returning to FIG. 1, the grain extraction unit 111 is set by the grain length setting unit 109 from the extraction start position set by the extraction start position setting unit 107 (the extraction start position after randomization except at the beginning of the audio file). A plurality of grains are extracted from the audio file by the phase set by the overlap phase setting unit 110 with the set grain length. Grain extraction is simply an operation of designating the start position and end position of digital data on an audio file and copying it to a work area (memory), and is instantaneously performed regardless of the reproduction time.

  Next, the reproduction / window function processing unit 112 performs reproduction based on the pitch given from the controller operation determination unit 102 for each extracted grain (speed of reproduction at a length obtained by dividing the grain length by the pitch. Pitch “1”). In the case of recording at the same speed as in recording and double speed in the case of pitch “2”, etc., and synthesizing by performing window function processing for multiplying a predetermined window function for each grain.

  FIG. 9 is a diagram showing an example of grain extraction and reproduction. When the grain length is 100 [msec], (a) is tempo “1” and pitch “1”, (b) is tempo “0.5”. In the case of the pitch “2”, (c) is the case of the tempo “2” and the pitch “2”. Note that randomization and overlap of the extraction start position are omitted so as not to cause difficulty in the drawing.

  In (a), the locator value matches the actual processing time because of the tempo “1”, and because the pitch is “1”, the extracted grain is played back at the same speed as when recording, and the grain is continuously extracted. And reproduction is performed.

  In (b), since the tempo is “0.5”, the processing time is 0 to 2000 [msec] with respect to the locator value 0 to 1000, and because the pitch is “2”, the extracted grain is twice the recording time. (Reproduction grain length = length of window function = 100 [msec] ÷ 2 [pitch] = 50 [msec]), the locator value is 25 from the first grain extraction, and is superimposed on the previous grain. Then, the next grain is taken out and played back. That is, the grain to be reproduced first is taken from 0 to 100 on the locator, and the grain to be reproduced next (after 50 [msec]) is taken from 25 to 125 on the locator.

  In (c), because the tempo is “2”, the processing time is 0 to 500 [msec] with respect to the locator value 0 to 1000, and the grain extracted for the pitch “1” is reproduced at the same speed as the recording. The locator value is 200 from the previous grain extraction, and the next grain is extracted and reproduced after a blank period from the previous grain. In this case, the data of the audio file is lost, but the grain length is short, so the influence on the sound quality is small. If there is an influence, the grain length is controlled in accordance with the tempo and pitch, so that it can be controlled so as to fall within a predetermined sound quality.

  Note that the tempo / pitch combination (not shown) has one of the above-described patterns (a) to (c), although the relationship between the locator value and the actual processing time changes. As a special case, one grain is normally reproduced only once, but when the tempo is “0”, the extraction start position is not updated and the same grain is repeatedly reproduced.

  Although the case where the locator moves in the positive direction is illustrated, if the locator value is decreased at a speed corresponding to the tempo, it can be set to move in the negative direction. In this case, reverse playback is performed at a speed corresponding to the tempo.

  In addition, an example of controlling the grain length and / or randomization value by tempo and pitch has been described, but the control table (108) shows the correlation between the portion (position) of the music to be extracted and the grain length and / or randomization value. It is also possible to set the grain length and / or randomize value according to the position of the reference (locator value) of the extraction start position.

  In the above-described embodiment, the game apparatus 1 including the granular synthesis processing unit 105 and the controller 2 are separate bodies and perform wireless communication with each other. However, this may be wired communication or integrated as a game. The device can also serve as a controller.

  The present invention has been described above by the preferred embodiments of the present invention. While the invention has been described with reference to specific embodiments, various modifications and changes may be made to the embodiments without departing from the broad spirit and scope of the invention as defined in the claims. Obviously you can. In other words, the present invention should not be construed as being limited by the details of the specific examples and the accompanying drawings.

It is a figure which shows the structural example of the game device concerning one Embodiment of this invention. It is a figure which shows the structural example of an acceleration change interpretation part. It is a figure which shows the example of an acceleration change. It is explanatory drawing of a locator and taking-out position update. It is a figure (the 1) which shows the example of a control table. FIG. 10 is a second diagram illustrating an example of a control table. It is a figure which shows the example of an overlap. It is a figure which shows the example of a window function. It is a figure which shows the example of grain extraction and reproduction | regeneration.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Game device 101 Wireless communication part 102 Controller operation determination part 103 Acceleration change interpretation part 103a X direction acceleration acquisition part 103b Y direction acceleration acquisition part 103c Z direction acceleration acquisition part 103d Threshold processing part 103e Threshold processing part 103f Threshold processing part 103g Vibration period Detection unit 103h Vibration period detection unit 103i Vibration period detection unit 103j Rotation period detection unit 103k Rotation throw detection unit 103l Tempo / pitch change unit 104 Game control unit 105 Granular synthesis processing unit 106 Extraction start position update unit 107 Extraction start position setting unit 108 Control table 109 Grain length setting unit 110 Overlap phase setting unit 111 Grain extraction unit 112 Playback / window function processing unit 113 Window function 0 value determination unit 114 Grain length update unit 115 Memory 16 Media 117 media read unit 118 microphone 119 A / D converter 120 effector portion 121 panning unit 122 effector portion 123 volume changing unit 124 D / A converter section 125 speaker 2 controller 21 operation switch 22 three-dimensional acceleration sensor 23 wireless communication unit

Claims (5)

An extraction start position setting unit for setting an extraction start position for extracting grains from the audio file;
A grain length setting unit for setting the grain length of the grain to be extracted;
A grain take-out unit for taking out grains of the grain length from the take-out start position on the audio file;
A playback unit that plays back the extracted grains according to a set pitch;
A pickup start position update unit that gives a reference position on the audio file according to the set tempo when the reproduction of the grain is completed, and instructs the pickup start position setting unit to update the pickup start position ;
The extraction start position setting unit obtains a randomized value based on the current tempo and / or pitch from the control table, and randomly determines the extraction start position based on the randomized value before and after the reference position on the audio file. audio reproducing apparatus according to claim <br/> to make. The audio playback device according to claim 1 ,
A grain length update unit for instructing the grain length setting unit to update the grain length when the reproduction of the grain is completed;
The grain length setting unit acquires a grain length based on a current tempo and / or pitch from a control table. In the audio | voice reproduction apparatus as described in any one of Claim 1 or 2 ,
The grain extraction unit extracts a plurality of grains having different extraction start positions based on a predetermined overlap phase,
The audio reproduction apparatus characterized in that the reproduction unit performs synthesis after multiplying each grain by a predetermined window function. The audio playback device according to claim 3 ,
An audio reproducing apparatus comprising a window function zero-value determining unit that detects completion of grain reproduction when the window function reaches zero. The computers that make up the equipment,
An extraction start position setting means for setting an extraction start position for extracting grains from the audio file;
Grain length setting means for setting the grain length of the grain to be extracted,
Grain extraction means for extracting grains of the grain length from the extraction start position on the audio file;
Reproduction means for reproducing the extracted grain according to a set pitch,
When the reproduction of the grain is completed, a reference position on the audio file corresponding to the set tempo is given to function as an extraction start position update means for instructing the update of the extraction start position ,
The extraction start position setting means obtains a randomized value based on the current tempo and / or pitch from the control table, and randomly sets the extraction start position based on the randomized value before and after the reference position on the audio file. let <br/> audio reproduction control program.

Images