JP4940875B2 - Audio data compression method and apparatus, and decompression method and apparatus - Google Patents

Description

  The present invention relates to an audio data compression method and apparatus, and a decompression method and apparatus mainly used for a pachinko machine or a game machine.

In pachinko machines, game machines, etc., audio data for back music is compressed and stored in a ROM, and the ROM is set in a game machine or the like. When the game machine is used, the data in the ROM is read and decompressed during music playback, and back music is generated. In playback, the same phrase is played back repeatedly.
In the audio data compression described above, an irreversible compression method with a high compression rate such as MP3, AAC (Advanced Audio Coding), WMA (Windows (registered trademark) Media Audio) or the like is used. However, since this compression method with a high compression rate performs compression / expansion processing with a large number of samples, for example, 1024 samples as one unit, the decoding latency from the start of the expansion processing until the decompressed audio data is obtained. For this reason, in the case of repeated playback, the musical tone is temporarily stopped after the musical tone has been repeated and before the musical tone starts from the repetition point.

  Therefore, conventionally, in order to eliminate this temporary stop of the musical sound, for example, audio data having a predetermined length from the repetition point by a compression method which has low latency (decoding delay is small) such as ADPCM but does not require time for expansion. In the case of repeated playback, first, the data with this low latency is expanded and played back, and during that time, the data with high compression rate is expanded. Instead, music was played back using high compression rate data.

  By the way, in the above processing, if a musical tone based on low latency data and a musical tone based on high compression rate data are simply connected, click noise occurs at the connection point. Therefore, in order to eliminate this click noise, it is desirable to perform cross-fade processing that gradually attenuates the musical sound based on the low latency data at the connection point while gradually increasing the musical sound based on the high compression rate data.

Conventionally, as a method of the crossfade processing, there is a method of separately preparing high compression rate data that has been subjected to crossfade processing of a predetermined length. However, this method has the disadvantages that the data amount is increased correspondingly and the compression rate is lowered. As another method of crossfade processing, there is a method of incorporating a crossfade circuit into an audio data expansion circuit. However, this method has a drawback that the hardware is increased, and when a DSP (digital signal processor) is used, there is a disadvantage that the number of steps is increased.
Patent Documents 1 to 3 are known as conventional compression / decompression apparatuses that repeatedly reproduce audio data.
JP 2002-341896 JP 2004-348055 A JP 2005-283944 JP

  The present invention has been made in consideration of the above circumstances, and its purpose is to perform cross-fading processing without preparing audio data for cross-fading and without providing a cross-fading circuit. An object of the present invention is to provide an audio data compression method and apparatus and an expansion method and apparatus.

The present invention has been made in order to solve the above-mentioned problems, and the invention according to claim 1 is a method for compressing compressed data by a compression method with a high compression ratio and writing a packet obtained by the compression into a memory. and first step, from among the packets in said memory, the preset during playback point belongs packet and a second step of detecting the next following packet, the packets detected at said second step From the third step of decoding to obtain decoded data, from the compressed data corresponding to the intermediate reproduction point to the beginning position of the decoded data, and the compressed data corresponding to the predetermined portion from the beginning of the decoded data The fourth step of compressing the data obtained by subtracting the predetermined portion from the beginning of the decoded data by a compression method with low latency An audio data compression method is characterized by having a.

According to the second aspect of the present invention, the first compression is performed in which the data to be compressed stored in the first memory is read out, compressed by a compression method with a high compression ratio, and a packet obtained by the compression is written in the second memory. means, among the packets in the second memory, decodes a detector during playback point set in advance to detect the next following packet belongs packet, the detected packet in the detection means A decoder that outputs decode data; a first block of compressed data corresponding to the intermediate reproduction point to the start position of the decode data; and a second copy corresponding to a predetermined portion from the start of the decode data. Each block of compressed data is read from the first memory, and the first block of data to be compressed is compressed by a compression method using low latency. And a second compression means for compressing the data obtained by subtracting the decoded data from the block of the second compressed data by a compression method with a low latency. .

  According to a third aspect of the present invention, audio data is read out and decompressed from a memory in which compressed data with a high compression ratio and compressed data with low latency divided into a first part and a second part are stored. In the decompression method, a first step of sequentially reading out and decompressing the compressed data with the high compression rate from the memory and outputting the decompressed data, and a reproduction stop point in which the decompression process in the first step is set in advance The second step of reading the first portion of the compressed data with the low latency from the memory, decompressing and outputting the decompressed data, and the high compression from a specific position set in the memory A third step of reading and decompressing the compressed data by rate, and a second part of the compressed data by the low latency from the memory. And a fourth step of adding the decompressed data and the decompressed data obtained in the third step and outputting the addition result as decompressed data. Data decompression method. Is

According to a fourth aspect of the present invention, audio data is read and decompressed from a memory in which compressed data with a high compression ratio and compressed data with low latency divided into a first portion and a second portion are stored. In the decompression device, first means for sequentially reading out and decompressing the compressed data with the high compression rate from the memory and outputting the decompressed data, and a reproduction end point at which decompression processing in the first means is preset The second means for reading the first portion of the compressed data with the low latency from the memory, decompressing and outputting the decompressed data, and the high compression from a specific position set in the memory A third means for reading and decompressing the compressed data by the rate, and reading and decompressing the second portion of the compressed data by the low latency from the memory; Adds the decompressed data obtained and its decompressed data by said third means, a voice data expansion apparatus, characterized in that it comprises a fourth means for outputting the addition result as the decompressed data.

  According to the present invention, when the music data compression / decompression apparatus performs halfway reproduction of the music, the crossfade processing is performed without preparing the audio data for crossfading and without providing the crossfade circuit. There is an effect that can.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an audio data compression apparatus according to an embodiment of the present invention. In this figure, reference numeral 1 is a memory in which audio data (PCM data) before compression is stored, 2 is a control circuit, and 3 is a high compression ratio such as MP3, AAC, WMA, etc., but the pronunciation latency at the time of decoding is A first encoder that compresses large audio data, 4 is a second encoder that compresses audio data with low latency such as ADPCM, DPCM, LPC, uncompressed, but low latency during decoding, and 5 stores compressed data And 6 is a decoder for decompressing the audio data compressed by the first encoder. The compressed data in the memory 5 is generated into a bit stream by the bit stream generation circuit, packetized, and written into the ROM.

Next, the operation of the above-described audio data compression apparatus will be described with reference to FIGS.
First, the control circuit 2 sequentially reads the audio data before compression from the memory 1 and outputs it to the first encoder 3. The first encoder 3 compresses the audio data by a compression method with a high compression rate, and writes the compressed data in the memory 5 (step Sa1). When all the data in the memory 1 is compressed by the first encoder and written in the memory 5, the control circuit 2 next detects the packet to which the midway reproduction point belongs (step Sa2). The intermediate playback point is a point at which, for example, in the case of repeated playback, the next playback is started after the music has been played once, and is set in the control circuit 2 in advance.

Hereinafter, this block detection process will be described with reference to FIGS.
In FIG. 3, P1, P2,... Are compressed packets (number of samples = 2048). When these packets P1, P2,... Are sequentially decoded (expanded), blocks B1, B2,... Are obtained corresponding to the packets P1, P2,. Each packet is decoded and stored in a buffer within the time required for 1024 samples to be played at the sampling rate. In addition, in the case of a compression method in which blocks are divided such as MP3 and AAC, a window function is applied so that the blocks have an overlap of 1024 samples and the first and last samples of the block are 0. As shown in the figure, each decoded block is a curve with a curved envelope. The decoded data is obtained by adding the data of adjacent blocks. For example, the 0th to 1023th samples are obtained by adding the blocks B1 and B2, and the 1024th to 2047th samples are obtained by adding the blocks B2 and B3. The same applies hereinafter.

  Next, FIG. 4 is a flowchart of processing for detecting a packet for halfway reproduction. Now, let the playback point A (FIG. 3) on the way be the 2500th sample from the start point. First, the control circuit 2 reads the first packet P1 from the memory 5 (step Sb1). Next, the sample number when the read packet is decoded is detected. In the case of the packet P1, the block B1 is obtained by decoding. However, for the 1024 samples in the first half, no sample is generated because there is no decoded sample of the adjacent packet to be added. Next, it is checked whether or not the sample number 2500 at the midway reproduction point A has passed (step Sb3). In this case, the determination result is “NO”, and the process returns to Step Sb1.

Next, the control circuit 2 reads the second packet P2 from the memory 5 (step Sb1). Next, the sample number when the read packet is decoded is detected (step Sb2). In this case, sample numbers 0 to 1023 are detected. Next, it is checked whether or not the sample number 2500 at the midway reproduction point A has passed (step Sb3). In this case, the determination result is “NO”, and the process returns to Step Sb1. Thereafter, the same processing is repeated. When the packet P5 is read in step Sb1, sample numbers 3072 to 4095 are detected in step Sb2, the determination result in step Sb3 is “YES”, and the flow proceeds to step Sb4. In step Sb4, the next packet number, that is, the packet number P6 is written in the packet number register inside the control circuit 2.

The above is the process of step Sa2 in FIG. Next, when proceeding to step Sa3, it is assumed that the packet before the packet detected at step Sa2 was data 0, and forcibly decoding until normal decoding becomes possible.
That is, in the above example, since the packet P6 (block B6) is detected in step Sa2, the control circuit 2 reads the packet P6 from the memory 5 and outputs it to the decoder 6 in step Sa3. The decoder 6 decodes the packet P6 and returns it to the control circuit 2. The packets after the packet P7 can be normally decoded at the time of reproduction, and the decoding process by the decoder 6 is not performed.

  Next, when proceeding to step Sa4, the control circuit 2 reads the uncompressed data from the memory 1, and calculates the difference between the read data and the decoded data. That is, in the case of the above example, the control circuit 2 reads the 2500th to 5119th samples from the memory 1, subtracts “0” for the 4095th sample from the 2500th sample, and starts from the 4096th sample. For the 5119th sample, the decoded data (data of block B6) is subtracted. Next, when proceeding to step Sa5, the control circuit 2 outputs the data obtained by the above subtraction to the second encoder 4. The second encoder 4 compresses the data by a low latency compression method and writes the compressed data in the memory 5. As a result, low latency compressed data indicated by the symbol AD in FIG.

  As described above, the audio data compression apparatus according to the above embodiment compresses and writes all data in the memory 1 to the memory 5 by a compression method with a high compression rate, and also low-latency data having a predetermined length from the playback point A on the way. Is prepared in the memory 5. This is because when the playback returns to point A, the high compression rate data cannot be decoded immediately, so the low latency data is played back for a while and the high compression rate data playback is in time. This is because data compression with a high compression rate is performed again thereafter. In the case of the example of FIG. 3, the packet P5 cannot be decoded in time, and the decoded data is obtained from the 4096th sample according to the decoding of the packet P6.

Therefore, it is necessary to prepare low latency data for the 2500th to 4095th samples. Even if decoded data is obtained from the 4096th sample by decoding the packet P6, there is no sample to be added for the 4096th to 5119th samples. Audio data cannot be obtained. Therefore, the data of the decoded block B6 is subtracted from the 4096th to 5119th samples before compression, and the data obtained by the subtraction is compressed by the low latency compression method and written to the memory 5. At the time of reproduction, the low latency data is decoded and used for the 2500th to 4095th samples, and the high compression rate data and the low latency data are decoded for the 4096th to 5119th samples. Data obtained by decoding the data is added and used. After the 5120th sample, data with a high compression rate is decoded and used.

Next, an audio data expansion device that expands data compressed by the compression device described above will be described.
FIG. 5 is a block diagram showing the configuration of the audio data decompression apparatus. In this figure, 11 is a ROM in which compressed audio data is written, 12 is a control circuit, 13 is a first decoder for decompressing data of the first compression method (high compression rate compression method), and 14 is a second decoder. A second decoder that decodes data in a compression scheme (low latency compression scheme), 15 an adder that adds the decoded data by the first decoder and the decoded data by the second decoder, and 16 an output of the first decoder, A selector that selects and outputs one of the output of the second decoder and the output of the adder, 17 is a buffer memory in which the output of the selector is written, 18 is a D / A that converts the output of the buffer memory 17 into an analog signal It is a converter.

Next, the operation of the above-described audio data expansion device will be described with reference to the flowcharts shown in FIGS.
First, the control circuit 12 reads the first packet P <b> 1 (see FIG. 3) from the ROM 11 and outputs it to the first decoder 13. The first decoder 13 decodes the packet P1 within the time required for the 1024 samples to be sounded according to the sampling rate, and stores it in the first buffer (not shown) as the block B1 (FIG. 3). . Next, the control circuit 12 reads the second packet P2 and outputs it to the first decoder 13. The first decoder 13 decodes the packet P2 and stores it as a block B2 in a second buffer (not shown). Next, the first decoder 13 generates the 0th to 1023 decoded samples by adding the data after 1024 samples of the block B1 and the data of 0 to 1023 samples of the block B2, and buffers them via the selector 16 While writing to the memory 17, the control circuit 12 reads the third packet P 3 and outputs it to the first decoder 13. The first decoder 13 decodes the packet P3 and stores it as a block B3 in a third buffer (not shown). The data written in the buffer memory 17 is read by the clock pulse of the sampling frequency, converted into an analog signal by the D / A converter 18, and output to the speaker. Thereby, a musical sound is generated (steps Sc2 and Sc3).

  Thereafter, the control circuit 12 sequentially reads the packets P3, P4... From the ROM 11 and outputs them to the first decoder 13. As a result, a musical tone based on the packets P3, P4... Is generated. Each time the control circuit 12 outputs a packet to the first decoder, it determines whether or not the reproduction stop point set in advance has been reached (step Sc1 in FIG. 6). If the determination result is “NO”, the process proceeds to steps Sc2 and Sc3.

  Next, when the reproduction stop point is reached, the first decoder 13 is instructed to clear the previously read data (step Sc4). Next, the data reading position by the first compression method is set to the decoding restart position determined at the time of encoding (step Sc5). In the case of the example of FIG. 3, the data reading position by the first compression method is set to the position of the packet P6. This position is written in the ROM 11 in advance.

  Next, the control circuit 12 reads the data of the second compression method from the ROM 11 (step Sc6). Then, the portion to be reproduced only with the data of the second compression method (in the example of FIG. 3, the 2500th to 4095th samples) is output to the second decoder 14. The second decoder 14 decodes this data and writes it into the buffer memory 17 via the selector 16 (step Sc7).

  Next, the control circuit 12 reads the data of the first compression method from the above-described decoding restart position of the ROM 11 and outputs it to the first decoder 13 (step Sc8). The first decoder 13 decodes the data and outputs it to the adder 15 (step Sc9). Further, the control circuit 12 reads from the ROM 11 data (the 4096th to 5119th samples in the example of FIG. 3) to be added to the first compression method data with the second compression method data, and sends it to the second decoder 14. Output. The second decoder 14 decodes the data and outputs it to the adder 15 (step Sc10). The adder 15 adds the outputs of the first and second decoders 13 and 14 and writes them to the buffer memory 17 via the selector 16 (step Sc11). The above addition processing is performed until the end of the data of the second compression method. When the data of the second compression method is completed, musical tone reproduction using the data of the first compression method is performed thereafter (step Sc12).

  As described above, according to the above-described embodiment, the cross-fade processing is performed using the window function used in the frequency conversion preprocessing (blocking) in the compression method with a high compression ratio (FIG. 3). Sample numbers 4096 to 5120). As a result, the crossfade process can be performed without preparing audio data for crossfade and without providing a crossfade circuit.

In the above embodiment, the playback stop point is set as the end point of the block. However, the present invention is not limited to this, and the playback stop may be performed in the middle of the block. In this case, the determination of the reproduction abort is made in the first decoder.
In the above description, repeated reproduction is given as an example. However, the above embodiment can also be applied to a case where reproduction is performed while skipping part of the middle.
In the above-described embodiment, the concept of compressed data with a low latency includes uncompressed data.

  The present invention is used for generating back music in a pachinko machine or a game machine.

It is a block diagram which shows the structure of the audio | voice data compression apparatus by one Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the audio | voice data compression apparatus. It is explanatory drawing for demonstrating operation | movement of the audio | voice data compression apparatus. It is a flowchart which shows the process of step Sa2 in FIG. 2 in detail. It is a block diagram which shows the structure of the audio | voice data pressure expansion apparatus by one Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the audio | voice data expansion | extension apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Memory, 2 ... Control circuit, 3 ... 1st encoder, 4 ... 2nd encoder, 5 ... Memory, 6 ... Decoder, 11 ... ROM, 12 ... Control circuit, 13 ... 1st decoder, 14 ... 2nd Decoder, 15 ... adder, 16 ... selector, 17 ... buffer memory, 18 ... D / A converter.

Claims (4)

A first step of compressing data to be compressed by a compression method with a high compression ratio, and writing a packet obtained by the compression to a memory;
From among the packets in the memory, a second step of during playback point set in advance to detect the next following packet belongs packet,
A third step of decoding the packet detected in the second step to obtain decoded data;
The compressed data corresponding to the intermediate playback point to the start position of the decoded data and the compressed data corresponding to the decoded data from the start to a predetermined portion are subtracted from the start of the decoded data to the predetermined portion. A fourth step of compressing the obtained data by a compression method with low latency;
An audio data compression method characterized by comprising: First compression means for reading the compressed data stored in the first memory, compressing the compressed data by a compression method with a high compression ratio, and writing the packet obtained by the compression to the second memory,
From among the packets in the second memory, the detecting means during playback point set in advance to detect the next following packet belongs packet,
A decoder that decodes the packet detected by the detection means and outputs decoded data;
A first block of compressed data corresponding to the intermediate reproduction point to the beginning position of the decoded data and a second block of compressed data corresponding to the predetermined portion from the beginning of the decoded data are respectively Read from the memory of 1 and compress the block of the first compressed data by a compression method using low latency, and subtract the decoded data from the block of the second compressed data to obtain the low latency A second compression means for compressing by the compression method according to
An audio data compression apparatus comprising: In an audio data decompression method for reading and decompressing data from a memory in which compressed data with a high compression rate and compressed data with low latency divided into a first part and a second part are stored.
A first step of sequentially reading and decompressing the compressed data with the high compression rate from the memory, and outputting the decompressed data;
When the decompression process in the first step reaches a preset reproduction stop point, the first portion of the compressed data with the low latency is read from the memory, decompressed, and the decompressed data is output. And the steps
A third step of reading and decompressing the compressed data with the high compression ratio from a specific position set in the memory;
Reads and decompresses the second portion of the compressed data with the low latency from the memory, adds the decompressed data and the decompressed data obtained in the third step, and outputs the addition result as decompressed data A fourth step to
An audio data decompression method comprising: In an audio data decompression device that reads and decompresses data from a memory in which compressed data with a high compression rate and compressed data with low latency divided into a first part and a second part are stored.
First means for sequentially reading and decompressing the compressed data with the high compression rate from the memory, and outputting the decompressed data;
When the decompression process in the first means reaches a preset playback stop point, the first portion of the compressed data with the low latency is read from the memory, decompressed, and the decompressed data is output. Means of
A third means for reading and decompressing the compressed data with the high compression ratio from a specific position set in the memory;
Reads and decompresses the second portion of the compressed data with the low latency from the memory, adds the decompressed data and the decompressed data obtained by the third means , and outputs the addition result as decompressed data A fourth means to
An audio data decompressing device comprising:

Images