JP2001312276A - Time-base compressing/expanding reproducing device for audio waveform data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an audio waveform signal which is compressed and expanded along the time base accurately by following cycle information on a playing tempo, etc. SOLUTION: This device has a storage means which stores audio waveform data representing an audio waveform signal, a reproduced waveform signal generating means which generates a reproduced waveform signal by reading the audio waveform data out of the storage means to output it, a position data generating means which generates position data indicating positions on the stored audio waveform data at specific cycle intervals, a cycle information generating means which generates desired cycle information, and a waveform generation position control means which performs control over a jump of the readout position of the audio waveform data to positions that the position data generated by the cycle information generation means indicate in synchronism with the desired cycle information generated by the cycle information generating means when the reproduced waveform signal is generated by the reproduced waveform signal generating means; and the audio waveform data are compressed or expanded along the time base corresponding to the cycles of the desired cycle information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for compressing / expanding audio waveform data on a time axis, and more particularly, to compressing and reproducing stored audio waveform data on a time axis or reproducing stored audio waveform data. The present invention relates to a time axis compression / expansion / reproduction apparatus for audio waveform data capable of expanding and reproducing waveform data on a time axis.

[0002] In this specification, "time axis compression / expansion" of audio waveform data refers to "compression on time axis" (time axis compression) of audio waveform data and "time axis compression / expansion" of audio waveform data. "Extend on" (time-axis extension).

[0003]

2. Description of the Related Art Conventionally, audio capable of reproducing compressed audio waveform data stored in advance on a time axis and expanding and reproducing audio waveform data stored in advance on a time axis. A time axis compression / expansion reproduction apparatus for waveform data is known.

FIG. 1 is an explanatory diagram of a process performed by such a conventional audio waveform data time axis compression / expansion reproduction apparatus.

In FIG. 1, the audio waveform data of the audio waveform signal having the original tone generation time Torg, which is stored from address 0 to the address Aend, is compressed / expanded on the time axis to change the tone emission time. It is shown.

First, a time axis compression process for compressing audio waveform data on a time axis will be described with reference to FIG.

Here, the case where the original tone generation time Torg is compressed on the time axis to the tone generation time Tcmp will be described. In this case, in a conventional time axis compression / expansion / reproduction apparatus for audio waveform data, a two-dot chain line in FIG. A time-to-address change characteristic shown in FIG. 1 is obtained, and reading of audio waveform data is controlled based on the change characteristic to obtain time-axis-compressed audio waveform data.

However, when the audio waveform data is read out and the audio waveform data is reproduced using the address change characteristic itself shown by the two-dot chain line, the reading speed of the audio waveform data (the increment of the read position of the audio waveform data) changes. (Ie, the reading speed of the audio waveform data is increased), and the pitch of the reproduced audio waveform data is changed to a high pitch.

In order to prevent this, in the conventional audio waveform data time axis compression / expansion / playback apparatus, when reading the audio waveform data, the audio waveform data is read out at the same speed as when the original audio waveform data was sampled. Reading is performed by jumping the read position of the audio waveform data to the address indicated by the time-to-address change characteristic indicated by the two-dot chain line for each predetermined time Tx, and the pitch of the reproduced audio waveform data is changed. The reading of the audio waveform data is controlled so as not to change high.

As a result, the read address of the audio waveform data changes as shown by the thick solid line drawn along the two-dot chain line in FIG. 1, and when viewed as a whole, the time shown by the two-dot chain line in FIG. This makes it possible to reproduce the changing audio waveform data.

That is, by reading out the audio waveform data at the address indicated by the bold solid line, it is possible to reproduce the audio waveform data compressed only on the time axis without changing the pitch.

In the actual processing, the audio waveform data read at each time Tx is connected by cross-fade processing, thereby preventing noise at the connection points.

Next, with reference to FIG. 1, a description will be given of the time axis expansion processing for expanding the audio waveform data on the time axis.

In this time axis expansion processing, as in the case of the time axis compression processing described above, the original tone generation time T
A description will be given of a case where the org is expanded in time axis to the sound generation time Tstr. In this case, the conventional time axis compression / expansion / playback apparatus for audio waveform data obtains the time-address change characteristic indicated by the one-dot chain line in FIG. Thus, the reading of the audio waveform data is controlled on the basis thereof to obtain the time-axis-compressed audio waveform data.

However, when the audio waveform data is read out and the audio waveform data is reproduced with the address change characteristic itself indicated by the dashed line, the reading speed of the audio waveform data (the amount of advance of the audio waveform data reading position) changes. (That is, the reading speed of the audio waveform data becomes slow), and the pitch of the reproduced audio waveform data changes to be low.

In order to prevent this, in the conventional audio waveform data time axis compression / expansion / playback apparatus, when reading the audio waveform data, the audio waveform data is read out at the same speed as when the original audio waveform data was sampled. Reading is performed by jumping the reading position of the audio waveform data to the address indicated by the time-to-address change characteristic indicated by the one-dot chain line for each predetermined time Tx, and the pitch of the reproduced audio waveform data is changed. The reading of the audio waveform data is controlled so as not to change low.

As a result, the read address of the audio waveform data changes as shown by the thick solid line drawn along the one-dot chain line in FIG. 1, and when viewed as a whole, the time shown by the one-dot chain line in FIG. This makes it possible to reproduce the changing audio waveform data.

That is, by reading the audio waveform data at the address indicated by the thick solid line, it is possible to reproduce the audio waveform data expanded only on the time axis without changing the pitch.

In the actual processing, the audio waveform data read at each time Tx is connected by cross-fade processing, thereby preventing the occurrence of noise at the connection point.

By the way, in the conventional audio waveform data time axis compression / expansion / reproduction apparatus, the sounding time after the time axis compression / expansion as described above (the sounding time Tcmp in the case of time axis compression, In this case, the tone generation time Tstr is directly given.) There is no problem since the time axis compression / expansion may be performed using the directly given tone generation time. When performing time axis compression / expansion processing, time axis compression / expansion processing may be performed using tempo information.

For example, original waveform data having a performance tempo value of 100 BPM as tempo information
When changing to the performance tempo value of PM, the processing of compressing the time axis by "100/120" from this performance tempo value has been performed.

However, in such a performance tempo, an accurate value may not be obtained in some cases, or the number of significant digits cannot be obtained so much. , "100/120"
This is the calculation process of performing the calculation of. ), There is a problem that it is difficult to obtain audio waveform data compressed and expanded on the time axis at an accurate performance tempo due to a calculation error.

[0023]

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems of the prior art, and has as its object to accurately follow periodic information such as performance tempo. It is an object of the present invention to provide a time axis compression / expansion reproducing apparatus for audio waveform data which can obtain audio waveform data which has been time axis compressed / expanded.

[0024]

In order to achieve the above object, a time axis compression / expansion / reproduction apparatus for audio waveform data according to the present invention jumps a reading position of audio waveform data in synchronization with period information. The axis compression / expansion processing of audio waveform data is performed.

Therefore, according to the audio waveform data time axis compression / expansion reproduction apparatus of the present invention, it is possible to perform the time axis compression / expansion processing of audio waveform data by accurately following the period information.

Further, according to the audio waveform data time axis compression / expansion reproduction apparatus of the present invention, even if the time axis compression / expansion processing is performed in real time, the time axis compression / expansion of the audio waveform data can be accurately followed in time. The processing will be executed.

Further, when the cycle information of the cycle corresponding to the reproduction tempo is used as the cycle information, it is possible to obtain an audio waveform signal which is compressed and decompressed on the time axis and accurately corresponds to the reproduction tempo.

Further, when the cycle information of the cycle corresponding to the reproduction tempo is used as the cycle information, the jump processing of the audio waveform data is performed in synchronization with the performance tempo. Since it occurs in synchronization with the performance tempo, unnaturalness is less noticeable, and a high-quality tone can be generated.

That is, according to a first aspect of the present invention, there is provided a storage means for storing audio waveform data representing an audio waveform signal, and reading out the audio waveform data stored in the storage means to reproduce a reproduced waveform signal. A reproduction waveform signal generating means for generating and outputting a signal; a position data generating means for generating position data indicating the audio waveform data stored in the storage means at predetermined intervals; and a cycle for generating desired cycle information In the information generating means and the generation of the reproduced waveform signal in the reproduced waveform signal generating means,
A waveform for controlling a jump of a read position of the audio waveform data to a position represented by the position data generated by the position data generation unit in synchronization with the desired period information generated by the period information generation unit. Generating position control means for compressing and expanding the audio waveform data on the time axis in accordance with the cycle of the desired cycle information.

Therefore, according to the first aspect of the present invention, the process of jumping the read position, which is the reproduction position of the audio waveform data, is performed according to the cycle information of the desired cycle. Time axis compression / expansion processing of audio waveform data that accurately follows information can be performed, and even when time axis compression / expansion processing is performed in real time, time axis compression / expansion processing is executed accurately following time changes Will be done.

The relationship between each component of the first aspect of the present invention and each configuration in the embodiment described below will be described. The "storage means" will be described later. The “reproduced waveform signal generating means” corresponds to “reproducing processing” in an embodiment described later, and the “position data generating means” corresponds to an embodiment described below. The "position data" corresponds to the "address created as a template" in the embodiment described later, and the "period information generating means" corresponds to the "template creation processing" in the embodiment described later. The “period information” corresponds to “sequencer 12”, and the “period information” corresponds to “tempo clock” in the embodiment described later. Control means "corresponds to" jump processing in the tempo follow-up process "in the embodiment to be described later.

According to a second aspect of the present invention, an audio waveform data representing an audio waveform signal and first position data representing a position on the audio waveform data at a desired period interval are stored. Means for reading out the audio waveform data stored in the storage means, generating and outputting a playback waveform signal, and a playback waveform signal generation means for outputting the playback waveform signal. Based on the first position data stored in the storage means, Second position data generating means for generating second position data representing a position on the audio waveform data having a shorter period interval than the first position data, period information generating means for generating desired period information, and the reproduced waveform signal In the generation of the reproduction waveform signal by the generation unit, the read position of the audio waveform data is generated by the period information generation unit. In synchronism with the desired period information, the second generated by the second position data generating means
Waveform generation position control means for controlling jumping to a position represented by the position data, wherein the audio waveform data is compressed and decompressed on a time axis in accordance with the cycle of the desired cycle information.

Therefore, according to the second aspect of the present invention, the position represented by the second position data generated based on the stored first position data in accordance with the cycle information of the desired cycle. Since the process of jumping the read position, which is the reproduction position of the audio waveform data, is performed, the time axis compression / expansion processing of the audio waveform data that accurately follows the desired period information can be performed, and the time axis can be processed in real time. Even if the compression / expansion processing is performed, the time axis compression / expansion processing is executed accurately following the time change.

The relationship between each component of the present invention described in claim 2 of the present invention and each configuration in the embodiment described later will be described. The above-mentioned "storage means" will be described later. The “first position data” corresponds to “recorder time for each bar” in the embodiment, and the “reproduced waveform signal generating means” corresponds to “reproduced waveform signal” in the embodiment described later. The "second position data generating means" corresponds to "template creation processing" in the embodiment described later, and the "second position data" corresponds to "template as a template" in the embodiment described later. The "period information generating means" corresponds to the "sequencer 12" in the embodiment described later, and the "period information" corresponds to the Corresponds to "tempo clock" in the embodiment that was the "waveform generation position controlling means"
Corresponds to “jump processing in tempo tracking processing” in the embodiment described later.

According to a third aspect of the present invention, there is provided the audio waveform data representing an audio waveform signal played according to a desired performance tempo, and the audio waveform data having a period interval corresponding to the desired performance tempo. Storage means for storing first position data representing an upper position; read-out audio waveform data stored in the storage means to generate and output a reproduction waveform signal; and storage means A second position data generating means for generating, based on the first position data, second position data representing positions on the audio waveform data having a shorter period interval than the first position data; In the generation of the reproduction waveform signal by the period information generation means for generating the corresponding period information and the reproduction waveform signal generation means, The reading position of the I Oh waveform data,
A waveform generation position for performing control to jump to a position represented by the second position data generated by the second position data generation unit in synchronization with the period information corresponding to the reproduction tempo generated by the period information generation unit. Control means for compressing and expanding the audio waveform data on a time axis in accordance with the reproduction tempo.

Therefore, according to the third aspect of the present invention, according to the cycle information corresponding to the reproduction tempo,
Since the process of jumping the read position, which is the reproduction position of the audio waveform data, to the position represented by the second position data generated based on the stored first position data is performed, the audio corresponding to the reproduction tempo exactly. The waveform data can be compressed and decompressed in the time axis, and unnatural musical sounds due to jump processing are generated in synchronization with the playback tempo, so that unnaturalness is not noticeable and high-quality musical sounds can be generated. Will be able to

The relationship between each component of the present invention described in claim 3 of the present invention and each configuration in the embodiment described below will be described. The “first position data” corresponds to “recorder time for each bar” in the embodiment, and the “reproduced waveform signal generating means” corresponds to “reproduced waveform signal” in the embodiment described later. The "second position data generating means" corresponds to "template creation processing" in the embodiment described later, and the "second position data" corresponds to "template as a template" in the embodiment described later. The "period information generating means" corresponds to the "sequencer 12" in the embodiment described later, and the "period information" corresponds to the Corresponds to "tempo clock" in the embodiment that was the "waveform generation position controlling means"
Corresponds to “jump processing in tempo tracking processing” in the embodiment described later.

According to a fourth aspect of the present invention, the automatic performance is performed in accordance with the set desired performance tempo, a tone signal is generated, and the first period corresponding to the desired performance tempo is generated. Automatic performance means for generating information and second cycle information having a cycle shorter than the first cycle information; automatically playing the automatic performance means in accordance with a desired performance tempo; and sampling an audio waveform signal inputted to the input means. Sampling means for storing in the storage means audio waveform data representing the audio waveform signal, and first position data representing the positional relationship between the audio waveform signal and the first period information; A desired playback tempo different from the desired performance tempo is set, the performance is automatically performed in accordance with the playback tempo, and the automatic playback stored in the storage means is performed. A reproduction waveform signal generating means for generating and outputting a reproduced waveform signal read out I Oh waveform data, based on the first position data stored in the storage means,
A second position data generating means for generating second position data representing a position on the audio waveform data having a shorter period interval than the first position data; The read position of the waveform data is determined by the second performance generated by the automatic performance means which is automatically performed according to the reproduction tempo.
Waveform generation position control means for performing control to jump to a position represented by the second position data in synchronization with the cycle information,
The audio waveform data is compressed and expanded on a time axis in accordance with the reproduction tempo.

Therefore, according to the fourth aspect of the present invention, the first period information stored in synchronization with the second period information generated by the automatic performance means automatically performing the reproduction in accordance with the reproduction tempo. Since the process of jumping the read position, which is the reproduction position of the audio waveform data, to the position represented by the second position data generated based on the position data is performed, the time axis compression and decompression of the audio waveform data exactly corresponding to the reproduction tempo is performed. The processing can be performed, and the unnatural musical sound due to the jump processing is generated in synchronization with the reproduction tempo, so that the unnaturalness is not conspicuous and a high-quality musical sound can be generated.

The relationship between each component of the invention described in claim 4 of the present invention and each configuration in the embodiment described later will be described. The “first cycle information” corresponds to “measure information” in the later-described embodiment, and the “second cycle information” corresponds to the later-described embodiment. "Sampling means" corresponding to the "tempo clock" in
Corresponds to the “recording process” in the embodiment described later,
The above-mentioned "first position data" corresponds to "recorder time for each bar", the above-mentioned "storage means" corresponds to "recorder memory" in an embodiment described later, and the above-mentioned "second position data generation means". "Corresponds to" template creation processing "in the embodiment described later, and the" second position data "corresponds to" address created as a template "in the embodiment described later, and the" reproduced waveform signal "described above. The "generation means" corresponds to "reproduction processing" in the embodiment described later, and the "waveform generation position control means" corresponds to "jump processing in tempo tracking processing" in the embodiment described later.

Here, in the invention according to any one of claims 2, 3 and 4, as in the invention according to claim 5, the storage means stores the audio data Tempo data representing the desired performance tempo is stored together with the waveform data, and the second position data generating means stores the second position data based on the first position data and the tempo data stored in the storage means. It may be generated.

The relationship between each component of the present invention described in claim 5 of the present invention and each component in the embodiment described later will be described. The above-mentioned "tempo data" will be described later. This corresponds to “tempo at the time of recording” in the embodiment.

Further, in the invention according to any one of claims 2, 3 and 4, as in the invention described in claim 6, the second position data generating means may include: Alternatively, the second position data may be generated by equally dividing the first position data.

[0044]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an apparatus for compressing and expanding audio waveform data on a time axis according to the present invention will be described in detail with reference to the accompanying drawings.

In the following description, for the purpose of facilitating the understanding of the present invention, the description of the technical contents to which the conventional technology can be applied is omitted, and the audio contents, which are the technical contents related to the implementation of the present invention, are omitted. Only the time axis compression / expansion processing of data will be described in detail.

First, FIG. 2 shows a block diagram of a system including an embodiment of a time axis compression / expansion / playback apparatus for audio waveform data according to the present invention.
The entire operation of the system shown in FIG. 2 is controlled by a microcomputer (not shown).

The system shown in FIG.
And a sequencer (automatic performance device including a sound source) 12.

Here, the recorder 10 comprises a sequencer 12
Supplies a tempo clock of "1/32" beats, bar information and a performance tempo value.

In this embodiment, the recorder 10 is assumed to be recorded in monaural for the sake of simplicity and ease of understanding. Of course you can.

Further, the system shown in FIG. 2 has an input section 14 for inputting an audio waveform signal which is an analog signal from outside the system, and an audio waveform signal which is an analog signal input to the input section 14 being a digital signal. An analog / digital converter (A / D) 16 for performing analog / digital conversion to audio waveform data, and a recorder 1
A memory 18 which functions as a storage means for storing a recording operation, a recording operation, a reproduction operation, a stop operation, a tempo setting operation, a performance pattern selection operation, and an audio waveform data selection operation. User Interface (U)
I) 20, a synthesizing unit 22 for synthesizing audio waveform data output from the recorder 10 and the sequencer 12,
A digital / analog converter (D / A) 2 for converting audio waveform data, which is a digital signal, synthesized by the synthesizing unit 22 into an audio waveform signal, which is an analog signal.
4, an amplifier 26 for amplifying the audio waveform signal output from the digital / analog converter 24, and an amplifier 26
And an output unit 28 for outputting the audio waveform signal amplified in the above to an external speaker (not shown) or the like.

The user interface unit 20 detects the operation of each operation element and performs control for performing each processing according to the operation of each operation element.

In this system, as will be described later with reference to a flowchart, the sequencer 1
Even if the audio waveform data is recorded by the recorder 10 in time with the accompaniment 2 and the tempo of the sequencer 12 is changed from the time of recording when reproducing the audio waveform data as the recorded information, the recorder 10 still outputs the audio waveform as the recorded information. The data can be expanded and compressed on the time axis to follow the sequencer 12.

In this system, the synthesizing unit 2
By the processing of 2, the reproduced sounds of the recorder 10 and the sequencer 12 are mixed and output.

Further, since the accompaniment of the sequencer 12 can be freely created later or replaced with another accompaniment pattern, the accompaniment to be played at the time of recording by the recorder 10 is together with the recorder 10 at the time of reproduction. It is not necessarily the same as the accompaniment pattern to be played.

When the recording operator of the user interface unit 20 is operated, the UI recording processing routine shown in the flowchart of FIG. 3 is started, and the performance pattern and the tempo setting operator selected by the performance pattern selecting operator are activated. Is detected and stored in the header section 42 of a recorder memory (described later) whose data structure is shown in FIG. At the same time, a recording process routine shown in the flowchart of FIG.
Enable the start of interrupt processing.

When the stop operator of the user interface unit 20 is operated, the interruption processing of the recording processing routine shown in the flowchart of FIG. 4 is prohibited.

Here, the UI shown in the flowchart of FIG.
The recording processing routine will be described in detail. When the UI recording processing routine is started, first, at step S30
In the process 2, it is determined whether or not a performance pattern to be automatically played by the sequencer 12 is selected by the performance pattern selection operator of the user interface unit 20, and whether or not the sequencer 12 automatically plays by the tempo setting operator of the user interface unit 20. It is determined whether or not a performance tempo has been set.

If it is not determined in step S302 that both the performance pattern and the performance tempo are set, it is determined that both the performance pattern and the performance tempo are set. Until this, the determination processing of step S302 is repeated.

If it is determined in step S302 that both the performance pattern and the performance tempo have been set, the process proceeds to step S304, where the performance pattern and performance tempo are shown in FIG. In addition to storing the data configuration in the header section 42 of the recorder memory, the recording process of the recorder 10 is started by permitting the start of the interruption process of the recording process routine shown in the flowchart of FIG.

Here, the recorder memory is set in a predetermined area of the memory 18, and FIG. 4 shows a data structure stored in the recorder memory. In the processing of step S304, The performance pattern is stored in the header section 42 of the recorder memory as "accompaniment pattern selection data", and the performance tempo is stored as "tempo at the time of recording".

When the process of step S304 is completed, the process proceeds to step S306, and it is determined whether or not the stop operation element of the user interface unit 20 has been operated.

If it is determined in step S306 that the stop operator has not been operated, the determination process of step S306 is repeated until it is determined that the stop operator has been operated.

If it is determined in step S306 that the stop operator has been operated,
Proceeding to the process of step S308, the interruption process of the recording process routine shown in the flowchart of FIG. 4 is prohibited, the recording process of the recorder 10 is stopped, and the UI recording process routine ends.

Next, the recording process routine shown in the flowchart of FIG. 4 will be described in detail. This recording process routine is a process in the recorder 10, and the activation of the interrupt process is permitted in the process of step S304 of the UI recording process routine. This is an interrupt process executed for each sampling period.

When the recording operator of the user interface unit 20 is operated, the recorder 10 executes the processing shown in the recording processing routine, but the sequencer 12 executes step S3.
Automatic performance (accompaniment) is started according to the performance pattern and performance tempo set in 04.

When the recording process routine shown in the flowchart of FIG. 4 is started, first, in the process of step S402, audio waveform data is received from the analog / digital converter 16 and stored in the sample waveform memory unit 46 of the recorder memory. A process for storing as "sampling data" is performed.

When the processing in step S402 is completed, the process proceeds to step S404, where the sequencer 12
It is determined whether or not the current measure value has changed based on the measure information supplied to the recorder 10 from.

The processing for determining whether or not the current measure value has changed will be described in detail. The determination processing refers to “the current measure” stored in the buffer memory described below. It is done.

Here, the “current bar” is the sequencer 1
If it is determined that the input information is the bar information during the processing of another interrupt routine which is started every time information is input from the second bar, the input bar information is changed to the "current" as shown in FIG. In the buffer memory.

If it is determined in step S404 that the current bar value has changed, the process proceeds to step S406, in which the recorder time when the current bar value changes, that is, the beginning of the bar, Is recorded in the recorder time section 44 of the recorder memory as the “recorder time for each bar”.

Here, the “recorder time” is a value that is incremented (incremented) by one in one sample period, and the start of recording is defined as “0”. use.

When the process of step S406 is completed, the process proceeds to step S408. If it is not determined that the current bar value has changed in the determination process of step S404, the process proceeds to step S408.
Jump to the processing of and proceed.

Then, in the process of step S408, the recorder time is incremented, and the recording process routine ends.

Next, when the reproduction operation element of the user interface section 20 is operated, the UI reproduction processing routine shown in the flowchart of FIG. 6 is started, and the audio waveform data selected by the operation of the audio waveform data selection operation element is activated. (Sampling data) is detected and set as audio waveform data to be reproduced, and a reproduction processing routine shown in the flowchart of FIG.
Enable the start of interrupt processing.

When the stop operation element of the user interface unit 20 is operated, the interruption processing of the reproduction processing routine shown in the flowchart of FIG. 7 is prohibited.

Here, the UI shown in the flowchart of FIG.
The reproduction processing routine will be described in detail. When the UI reproduction processing routine is started, first, at step S60
In the process 2, the recorder 10 is operated by the audio waveform data selection operator of the user interface unit 20.
As audio waveform data (sampling data) to be reproduced by the sample waveform memory unit 4 of the recorder memory.
It is determined whether or not the audio waveform data recorded in No. 6 has been selected.

If it is not determined in step S602 that audio waveform data has been selected, the determination process in step S602 is repeated until it is determined that audio waveform data has been selected. become.

If it is determined in step S602 that the audio waveform data has been selected, the flow advances to step S604 to start the interrupt processing of the reproduction processing routine shown in the flowchart of FIG. Then, the reproduction process of the selected audio waveform data in the recorder 10 is started.

When the process of step S604 is completed, the process proceeds to step S606, and it is determined whether or not the stop operation element of the user interface unit 20 has been operated.

If it is determined in step S606 that the stop operator has not been operated, the determination process of step S606 is repeated until it is determined that the stop operator has been operated.

If it is determined in step S606 that the stop operator has been operated,
Proceeding to step S608, the interruption processing of the reproduction processing routine shown in the flowchart of FIG. 7 is prohibited, the reproduction processing of the recorder 10 is stopped, and this UI reproduction processing routine is ended.

Next, the reproduction processing routine shown in the flowchart of FIG. 7 will be described in detail. This reproduction processing routine is processing in the recorder 10, and the activation of the interruption processing is permitted in the processing of step S604 of the UI reproduction processing routine. This is an interrupt process executed for each sampling period.

In this reproduction processing routine, an interruption process is performed for each sampling period while the sequencer 12 supplies a tempo clock of "1/32" beat, a current performance tempo value (BPM) and current bar information. Things.

Here, the "tempo clock", "performance tempo value", and "current bar" are the values of the input information during the processing of another interrupt routine which is started every time information is input from the sequencer 12. If it is determined that they are the tempo clock, the performance tempo value, and the bar information, as shown in FIG. 8, the input tempo clock is stored in the buffer memory as a "tempo clock", and the input performance tempo value is displayed as "performance tempo clock". This is obtained by storing the measure information in the buffer memory as "tempo value" and storing the inputted measure information in the buffer memory as "current measure".

When the reproduction processing routine shown in the flowchart of FIG. 7 is started, first, in the processing of step S702, the “tempo information at the time of recording (original tempo)” stored at the time of recording and the current sequencer 12 "Performance tempo value (current t
empo) ”to determine whether they match.

Note that "tempo information at the time of recording (origi
As the “nal tempo”, the performance tempo stored as “tempo at the time of recording” in the head section 42 of the recorder memory is used.

In the above-described determination processing in step S702, the “tempo information at the time of recording (original te
If the current tempo value is equal to the current performance tempo value (current tempo) from the sequencer 12, the process proceeds to step S704 to determine whether a tempo clock has arrived.

If it is determined in step S704 that the tempo clock has arrived,
The process advances to step S706 to determine whether or not the current measure value has changed based on the measure information supplied from the sequencer 12 to the recorder 10. The process of determining whether or not the current bar value has changed is performed with reference to the “current bar” stored in the buffer memory.

If it is determined in step S706 that the current bar value has changed,
The process advances to step S708 to output a request for a template creation request for the next bar. To create a template,
Based on the request for the "template creation request for the next bar", processing is performed by a template creation routine (step S730), which is another processing routine.

The processing for creating a template will be described later with reference to an explanatory diagram of the processing executed by the template creating routine shown in FIG.

When the processing in step S708 is completed, the flow advances to step S710 to set the tempo clock tempo_clock to “0” and set the beat number beat to “0”.

When the process of step S710 is completed, the process proceeds to a process of step S711.

On the other hand, if it is not determined in step S706 that the current bar value has changed, the flow advances to step S712 to increment the tempo clock tempo_clock.

When the processing in step S712 is completed, the flow advances to the processing in step S711.

In the process of step S711,
It is determined whether or not the tempo clock tempo_clock is 32.

If it is determined in step S711 that the tempo clock tempo_clock is 32, the flow advances to step S713.
The tempo clock tempo_clock is set to “0” and the number of beats beat is incremented. When the processing in step S713 ends, the flow advances to the processing in step S714.

On the other hand, in the process of step S711,
If it is determined that the tempo clock tempo_clock is not 32, the process proceeds to step S714.

If it is not determined in step S704 that the tempo clock has arrived, the process jumps to step S714.

In step S714, sampling data (audio waveform data) is read from the sample waveform memory unit 46 of the recorder memory and output according to a read pointer (readPtr) indicating a read position of the audio waveform data.

When the processing in step S714 is completed, the flow advances to step S716 to increment the read pointer (readPtr) and terminate the reproduction processing routine.

Note that, in the above, step S708
The template creation routine is executed in step S730 based on the request of “Request for template creation of next bar”. That is, the processing from step S704 to step S716 includes processing for creating a template. If the tempo does not change between recording and playback, the processing is actually performed by processing for creating a template. No template information is required.

However, when the tempo is changed by the operation of the tempo setting operator of the user interface section 20 during the performance, a process for creating a template is performed so that the tempo can be immediately followed. Will be kept.

On the other hand, in the determination processing of step S702, the “tempo information at the time of recording (original item)
po) "and the current" tempo tempo value (current tempo) "from the sequencer 12 are not determined to be the same, the process proceeds to step S718, and the" performance tempo value (current tempo) "
Is smaller than the maximum tempo value LimitBPM (the maximum tempo value LimitBPM is, for example, “100 (BPM)”).

If it is determined in step S718 that the “performance tempo value (currenttempo)” is smaller than the maximum tempo value LimitBPM, the process proceeds to step S720, and the clock mode clock_mode is set to “1/32”. Set the mode.

When the processing in step S720 is completed, the flow advances to the processing in step S724.

On the other hand, in the judgment processing of step S718, the "performance tempo value (current tempo)"
Is not determined to be smaller than the maximum tempo value LimitBPM, the process proceeds to step S722.
The clock mode clock_mode is set to the “1/16” mode.

When the processing in step S722 ends, the flow advances to the processing in step S724.

Here, in step S724, a tempo follow-up processing routine (flow chart shown in FIG. 9) is executed as a subroutine of this reproduction processing routine. The reproduction processing routine ends.

Next, the tempo tracking processing routine shown in the flowchart of FIG. 9 will be described. In this tempo tracking processing routine, the read pointer (readP)
tr) increment processing (read pointer (rea
dPtr) by one address. ), The audio waveform data is read without changing the reading speed of the audio waveform data.

However, when reproducing at a tempo different from the tempo at the time of recording, it is necessary to correct the read address so that it matches the time axis. Specifically, for example, it is necessary to perform a process of correcting on a one-dot chain line in FIG. 10 described later and a process of correcting on a two-dot chain line in FIG. 11.

Then, the tempo clock is received from the sequencer 12 and the read pointer (readPtr) is jumped to correct the read position of the audio waveform data on the one-dot chain line in FIG. 10 or on the two-dot chain line in FIG. (Step S916 described later)
reference).

However, the read pointer (readPt)
r) When the audio waveform data is read out by itself and an audio waveform signal compressed and expanded on the time axis is generated, the read address jumps by the above-mentioned correction processing (the processing of jumping the read pointer), so that noise is generated.

Therefore, in order to prevent the occurrence of such noise, in this system, cross-fade processing is performed at the position where the address jumps.

That is, in the processing of step S918 described later, basically, the read pointer (readPt
r), the audio waveform data is read according to the read pointer (read) in the cross-fade portion.
The audio waveform data is read out so as to cross-fade slightly differently from Ptr).

Then, a crossfade routine (CrossFade Routine) of step S920, which will be described later, as a subroutine of the tempo follow-up processing routine
In the above, the audio waveform data read out is subjected to cross-fade processing to generate an audio waveform signal that is compressed and expanded on the time axis.

Hereinafter, the tempo following processing routine shown in the flowchart of FIG. 9 will be described in detail. First, in the processing of step S902, it is determined whether or not a tempo clock has arrived.

If it is determined in step S902 that the tempo clock has arrived,
Proceeding to step S904, it is determined whether or not the current measure value has changed based on the measure information supplied from the sequencer 12 to the recorder 10. The process of determining whether or not the current bar value has changed is performed with reference to the “current bar” stored in the buffer memory.

If it is determined in step S904 that the current bar value has changed,
The process advances to step S906 to output a request for a template creation request for the next bar. The template creation is processed by a template creation routine (step S930), which is another processing routine, based on the request for “a request for creating a template for the next bar”.

When the process of step S906 is completed, the process proceeds to step S908, where the tempo clock tempo_clock is set to “0” and the beat number beat is set to “0”.

When the processing in step S908 ends, the flow advances to the processing in step S911.

On the other hand, if it is not determined in step S904 that the current bar value has changed, the flow advances to step S910 to increment the tempo clock tempo_clock.

When the processing in step S910 ends, the flow advances to the processing in step S911.

In the process of step S911,
It is determined whether or not the tempo clock tempo_clock is 32.

If it is determined in step S911 that the tempo clock tempo_clock is 32, the flow advances to step S913.
The tempo clock tempo_clock is set to “0” and the number of beats beat is incremented. When the process in step S913 ends, the flow advances to the process in step S912.

On the other hand, in the process of step S 911,
If it is determined that the tempo clock tempo_clock is not 32, the process proceeds to step S912.

In the process of step S912,
It is determined whether the clock mode clock_mode is the “1/32” mode or the “1/16” mode.

In the determination processing in step S912, the clock mode clock_mode is set to “1/1”.
If it is determined that the mode is “6” mode, step S9
Proceeds to the process of 14, and the tempo clock tempo_clo
It is determined whether or not ck is an even number.

If it is determined in step S914 that the tempo clock tempo_clock is an even number, the flow advances to step S916.

On the other hand, in the judgment processing of step S912, the clock mode clock_mode is set to "1/3".
If it is determined that the mode is “2” mode, step S9
Jump to process 16 and proceed.

Here, in the processing of step S916, the number of beats beat and the tempo clock tempo_c
value of template in lock (templa
te [beat] [tempo_clock]) is set as the read pointer (readPtr), the tempo clock is received from the sequencer 12, the read pointer (readPtr) is jumped, and the read position of the audio waveform data is corrected. When the process in step S916 ends, the flow advances to the process in step S918.

On the other hand, when it is not determined that the tempo clock has arrived in the determination processing of step S902, or when it is not determined that the tempo clock tempo_clock is an even number in the determination processing of step S914. Then, the process proceeds to step S918.

In the process of step S918, a process of “reading out audio waveform data according to a read pointer (readPtr)” is executed as a subroutine of a tempo following process routine. In the process of step S918, basically, the read pointer (read
Ptr), the audio waveform data is read out, but the read pointer (rea)
dPtr) and audio waveform data is read out so as to cross-fade slightly differently.

When the processing in step S918 is completed, the flow advances to step S920 to execute a “crossfade routine” as a subroutine of a tempo tracking processing routine, and to read the read audio waveform data in a crossfade. Processing is performed to generate an audio waveform signal that has been compressed and expanded on the time axis.

When the processing in step S920 is completed, the process proceeds to step S922, in which the read pointer (readPtr) is incremented, and the reproduction processing routine ends.

The processing of “reading out audio waveform data in accordance with read pointer (readPtr)” in step S918 described above and step S92
0 "Cross fade routine (CrossFa
A known technique can be applied to the “de Routine)” process, and a detailed description thereof will be omitted.

Here, FIG. 10 is an explanatory diagram of the processing in the case where the reproduction is performed with the time axis extended (time axis extended reproduction).

That is, FIG. 10 shows the clock mode clo.
The tempo clock and the read pointer (r) when expanding the time axis when ck_mode is in the “1/16” mode
readPtr) is shown (thick solid line).

On the other hand, FIG. 11 is an explanatory diagram of a process in the case of reproducing the data by compressing the time axis (time-axis compression reproduction).

That is, FIG. 11 shows the clock mode clo.
Tempo clock and read pointer (r) during time axis compression when ck_mode is "1/16" mode
readPtr) is shown (thick solid line).

Here, the original tempo clock indicates the relationship between the tempo clock during recording and the address, and the reproduction tempo clock indicates the relationship between the tempo clock whose tempo has been changed (tempo clock during reproduction) and the address.

When the reproduction tempo changes, the timing of the clock changes. If the address position in the clock is corrected to the address position at the time of recording, the clock can follow the reproduction tempo.

That is, the read pointer (readPtr)
Jumps at the same time as the tempo clock arrives, but the actual read address is read in such a manner that the cross-fade section Tc overlaps.

Next, referring to FIG.
730 and the processing contents executed by the template creation routine in step S930 will be described.

In this system, in order to improve the efficiency of calculating the address at the time of recording for each tempo clock, a template composed of one bar is created in the template creation routine.

"Tempo_cl" shown in FIG.
“ock” is a value obtained by counting up the tempo clock of “1/32” beat from the sequencer 12, and is reset to “0” every beat. The address value used when the address is jumped and corrected is "a
(N) "(n = 0, 1, 2, 3,..., 29, 30 or
31).

Here, the current bar information is supplied from the sequencer 12, and when the current bar information is changed, a request for a template creation request for the next bar is issued to the template creation routine (step S708, step S708). Step S906).

The template for the current bar has already been prepared because a request for creation has been similarly given at the beginning of the previous bar.

That is, the template is a double buffer, and is alternately created and used.

Next, the address calculation for template creation will be described with reference to FIG. 12. First, the start address of the template is calculated from the recorder time T (n) for each measure recorded at the time of recording. calculate.

That is, template [0] [0] which is the start address of the first beat of the bar of the template is template [0] [0] = a0 = T (n).

The address of audio waveform data for each clock at the time of recording can be calculated if the tempo value at the time of recording is known.

The readout amount of audio waveform data per time interval of “1/32” beats is: readout amount of audio waveform data per time interval of “1/32” beats = 60 / tempo × l / Tf × 1/3
2 where tempo: tempo Tf = 1 / Fs: sampling frequency

Therefore, if the starting address of the performance is “0”, the address of an arbitrary clock is a (n) = (60 / tempo × l / Tf × l / 32)
× Cumulative clock can be calculated by the following formula.

In this system, since the recorder time T (n) (meas) at the beginning of a measure is recorded in the recorder time section 44 of the recorder memory at the time of recording, the start address of the measure = T (n) ) (Meas) a (n) = measure start address + (60 / tempo ×
1 / Tf × l / 32) × (cumulative tem from beginning of measure
po_clock).

As a result, the tempo value at the time of recording and the performance position on the time axis can be known in "1/32" units.
The address value of the audio waveform data at the performance time can be calculated.

As described above, even if the tempo of the accompaniment changes between recording and playback, it is possible to calculate and correct the read address of the audio waveform data synchronized with the playing time every "1/32" beat. become able to.

Although the above description has been made with reference to the clock of "1/32" beat, it is needless to say that a clock of an arbitrary interval such as "1/24" beat used in the MIDI signal can be used. It can be determined appropriately according to the conditions at the time of mounting. Apart from that, in this system, "1/32" beat and "1/16" beat
A beat is used properly (see FIGS. 7 and 13).

That is, in this system, a clock of "1/32" beat is received from the sequencer at the time of reproduction, and the clock is used as it is in the "1/32" beat mode. In the 16-beat mode, every other clock is used, and a clock having half the timing is used.

That is, in the examples shown in FIGS. 10 and 11, every other tempo clock of "1/32" beat is used, but in the example shown in FIG. 13, the tempo clock of "1/32" beat is used. Jump processing is performed for each clock (see "timing at which" jump processing "occurs" in FIG. 12).

Since the "1/16" beat mode processes in a rougher cycle than the "1/32" beat mode, there is an advantage that the system load is reduced.

Therefore, in the time axis compression processing for stopping the tempo, the load on the sequencer and the sound source increases, so
The "/ 16" beat mode is suitable.

On the other hand, in the time axis expansion processing for slowing down the tempo, the address is returned, so that if the tempo is extremely reduced, the sound quality tends to be delayed. For this reason, in the “1/32” beat mode, “l /
Since the sound quality can be improved by returning the address at a finer time interval than in the 16-beat mode, this system uses "1/1 /"
The mode is switched from the "6" beat mode to the "1/32" beat mode.

The above-described embodiment is described in the following (1)
It may be modified as shown in (4) to (4).

(1) In the above embodiment, the reading speed of audio waveform data (the increment of the read pointer (readPtr)) is the same as the writing speed, and the audio waveform data is incremented by one. However, it is a matter of course that the present invention is not limited to this, and when it is desired to change the pitch at the time of reproduction, the reading speed (the amount of advance of the read pointer (readPtr)) at the time of reproduction may be changed. That is, the higher the reading speed, the higher the pitch at the time of reproduction, and the lower the reading speed, the lower the pitch at the time of reproduction.

(2) In the above embodiment, one "tempo data at the time of recording" is stored for a series of sampling data. However, the present invention is not limited to this. When the "performance tempo value" is supplied from the automatic performance means (sequencer) and the tempo changes over time, the "tempo data at the time of recording" changed over time is stored. You may. Specifically, for example, “tempo data at the time of recording” may be stored for each bar.

(3) In the above-described embodiment, the template is created using the “tempo at the time of recording”. However, it is needless to say that the template is not limited to this.
For example, if a template is created by equally dividing one bar into 32 equal parts, the “tempo at the time of recording” may not be used.

(4) The above embodiments and the modifications shown in (1) to (3) above may be appropriately combined.

[0168]

Since the present invention is configured as described above, it is possible to obtain an audio waveform signal which is capable of obtaining an audio waveform signal which has been compressed and decompressed on a time axis by accurately following periodic information such as a performance tempo. And a time axis compression / expansion reproduction apparatus can be provided.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a process performed by a conventional audio waveform data time axis compression / expansion reproduction apparatus.

FIG. 2 is a block diagram of a system provided with an example of an embodiment of a time axis compression / expansion reproduction apparatus for audio waveform data according to the present invention.

FIG. 3 is a flowchart of a UI recording processing routine.

FIG. 4 is a flowchart of a recording process routine.

FIG. 5 is an explanatory diagram relating to a storage process of a “current measure” in a buffer memory.

FIG. 6 is a flowchart of a UI reproduction processing routine.

FIG. 7 is a flowchart of a reproduction processing routine.

FIG. 8 shows a “tempo clock” in a buffer memory,
FIG. 9 is an explanatory diagram relating to storage processing of a “performance tempo value” and a “current measure”.

FIG. 9 is a flowchart of a tempo tracking process routine.

FIG. 10 is an explanatory diagram of a process in a case where reproduction is performed with a time axis extended (time axis extended reproduction).

FIG. 11 is an explanatory diagram of a process in a case where a time axis is compressed for reproduction (time axis compression reproduction).

FIG. 12 is an explanatory diagram of processing contents executed by a template creation routine.

FIG. 13 is an explanatory diagram of a “1/32” mode.

[Explanation of symbols]

 Reference Signs List 10 recorder 12 sequencer 14 input section 16 analog / digital converter (A / D) 18 memory 20 user interface section (UI) 22 synthesizing section 24 digital / analog converter (D / A) 26 amplifier 28 output section 42 head section 44 Recorder time section 46 Sample waveform memory section

 ──────────────────────────────────────────────────続 き Continued on the front page F-term (reference)

Claims (6)

[Claims] A storage means for storing audio waveform data representing an audio waveform signal; a reproduction waveform signal generation means for reading out the audio waveform data stored in the storage means to generate and output a reproduction waveform signal; Position data generating means for generating position data indicating the audio waveform data stored in the storage means at predetermined periodic intervals; periodic information generating means for generating desired periodic information; and reproduction by the reproduced waveform signal generating means. In the generation of the waveform signal, the read position of the audio waveform data is set to a position represented by the position data generated by the position data generation unit in synchronization with the desired period information generated by the period information generation unit. Waveform generation position control means for performing jump control, and the audio waveform data Serial time warp extension and reproduction apparatus of the audio waveform data is to to time-axis compression and expansion corresponding to the period of the desired cycle information. 2. A storage means for storing audio waveform data representing an audio waveform signal and first position data representing a position on the audio waveform data at a desired periodic interval; and the audio waveform stored in the storage means. Reproduction waveform signal generation means for reading data to generate and output a reproduction waveform signal; and based on the first position data stored in the storage means, the audio waveform data having a shorter period interval than the first position data. Second position data generating means for generating second position data representing the above position; cycle information generating means for generating desired cycle information; and generating the reproduced waveform signal in the reproduced waveform signal generating means. The data reading position is synchronized with the desired period information generated by the period information generating means, Waveform generation position control means for performing control for jumping to a position represented by the second position data generated by the position data generation means, wherein the audio waveform data is stored in a time axis corresponding to a cycle of the desired cycle information. A time axis compression / expansion playback device for audio waveform data to be compressed / expanded. 3. An audio waveform data representing an audio waveform signal played according to a desired performance tempo and first position data representing a position on the audio waveform data at a period interval corresponding to the desired performance tempo are stored. A storage unit, a readout waveform signal generation unit that reads out the audio waveform data stored in the storage unit to generate and output a playback waveform signal, and based on the first position data stored in the storage unit, Second position data generating means for generating second position data representing positions on the audio waveform data having a shorter period interval than the first position data, and period information generating means for generating period information corresponding to a reproduction tempo; In the generation of the reproduced waveform signal in the reproduced waveform signal generating means, the read position of the audio waveform data is Waveform generation position control for performing control to jump to a position represented by the second position data generated by the second position data generation unit in synchronization with period information corresponding to the reproduction tempo generated by period information generation unit Means for compressing and expanding the audio waveform data on the time axis in accordance with the reproduction tempo. 4. An automatic performance corresponding to a set desired performance tempo, a tone signal is generated, and a first period information corresponding to the desired performance tempo and a second period information shorter than the first period information. Automatic performance means for generating period information; audio waveform data representing the audio waveform signal; performing automatic performance of the automatic performance means in accordance with a desired performance tempo, sampling an audio waveform signal input to the input means, Sampling means for storing in the storage means first position data representing the positional relationship between the audio waveform signal and the first period information; setting the automatic performance means to a desired reproduction tempo different from the desired performance tempo In addition to the automatic performance according to the reproduction tempo, the audio waveform data stored in the storage means is read out and the reproduction waveform signal is read out. Means for generating and outputting a reproduced waveform signal; and a second position representing a position on the audio waveform data having a shorter period interval than the first position data based on the first position data stored in the storage means. A second position data generating means for generating data; and a reproducing waveform signal generating means for generating a reproduced waveform signal, wherein the read position of the audio waveform data is generated by the automatic performance means for automatically performing according to the reproduction tempo. Waveform generation position control means for performing control to jump to a position represented by the second position data in synchronization with the second cycle information, and compressing / expanding the audio waveform data in accordance with the reproduction tempo. A time axis compression / expansion reproduction apparatus for audio waveform data. 5. The audio waveform data time-axis compression / expansion reproduction apparatus according to claim 2, wherein the storage means includes the desired performance together with the audio waveform data. An audio waveform storing tempo data representing a tempo, wherein the second position data generating means generates second position data based on the first position data and the tempo data stored in the storage means; A time axis compression / expansion playback device for data. 6. The time axis compression / expansion / playback apparatus for audio waveform data according to claim 2, wherein the second position data generating means is configured to output the first position data. A time axis compression / expansion reproduction apparatus for audio waveform data for generating second position data by equally dividing the interval.