JP2001189659A - Device and method for decoding compressed data - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent reproduced contents such as a decoded audio output, in the case of audio, from being broken, when decoding compression-encoded contents. SOLUTION: When decoding the data of compressed contents, a compressed data decoding device receives the stream of the data of these contents and applies data processing for decoding this stream and processed data are stored in a buffer, read at a prescribed sampling rate, made into analog signals and outputted. Furthermore, corresponding to the state of the compressed data decoding device or data processing conditions for decoding to the stream, a quality parameter is determined by a determining means 111 and corresponding to that quality parameter, the quality of data processing and analog signals are changed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the decoding of compressed data having a quality control function.
The present invention relates to a compressed data decoding device and a compressed data decoding method capable of converting digital audio data into an analog audio signal by performing quality control according to the situation at the time.

[0002]

2. Description of the Related Art There is MP3 as a compression / decoding technique for audio data which has a high data compression efficiency and can ensure good reproduction sound quality. Since MP3 can be realized by computer processing using software, it is a technology that is rapidly spreading in recent years. However, in some systems, in order to reproduce music without interruption of sound, a processor (CPU) is used. Requires an operation speed of about 100 [MHz].

[0003] The CPU speed required for the reproduction process may be about 50 [MHz] on average, but the required processing load varies depending on the data location.
z] or more processing power is required, and as a result, a CPU speed of about 100 [MHz] is required.

On the other hand, the MP3 player may be a general-purpose personal computer such as a personal computer as long as the MP3 software runs, but there is also a portable dedicated machine, which is about the size of a cigarette box. It basically comprises a CPU, a memory, an operation unit, an operation display, and an amplifier for amplifying an audio signal obtained by decoding, and is viewed through headphones or the like. In this case, since it is portable, the power supply for driving is a battery.

In any case, music data reproduction of MP3 is mainly performed by software processing by a CPU.

The power consumption of the CPU and its peripheral circuits is almost proportional to the operating frequency of the hardware. Therefore, in order to save power, the operating clock of the CPU and its peripheral circuits is reduced.

On the other hand, a portable device may be driven in a power saving mode due to the remaining capacity of a battery or the like. In such a case, the CPU and its peripheral circuits slow down the operation clock signal. To work, of course,
The decoding process slows down.

[0008] In order to reproduce music, it is necessary to convert the decoded audio data into an analog signal (D / A conversion) into an analog audio signal, which is supplied to a headphone or a speaker.

Here, the D / A conversion is 1 / 44.1 [k se
It is necessary to maintain a required D / A conversion speed such as c].

However, when the decoding processing speed becomes slow,
As a result, a "break" occurs in the audio output, for example, the decoding process cannot keep up with the sampling rate of the D / A conversion, and there is no data to be D / A converted, resulting in a silent state. Then, in the case of an audio signal, the occurrence of “interruption” is a large noise itself for a listener, and significantly impairs the quality of decoded speech.

[0011]

In a compressed audio decoding process for decoding compressed audio data, compressed and given audio data is converted into an original audio signal (audio signal).
In the case of an input stream that has a large load in the decoding processing part that decodes
The required D / A conversion speed (1 / 44.D.A.), Such as when the decoding device must always perform processing at low power.
1 [k sec]), the decoding processing speed may be delayed.

[0012] In such a case, as a result, a cause of generating "interruption" in the audio output is created. Then, in the case of audio, the occurrence of “interruption” is a large noise itself, which significantly impairs the quality of decoded speech.

Therefore, the object of the present invention is to:
First, even if the decoding processing speed is slow, if the content to be reproduced, for example, the content is audio data (audio data), the output of the reproduced audio signal is prevented from being interrupted so that the decoded audio is output. Another object of the present invention is to provide a compressed data decoding device and method having a voice quality control function capable of maintaining the quality of a compressed data.

Secondly, when the operation is performed in the power saving mode in the case of battery operation, the reproduced content, for example, if the content is audio data (audio data), the reproduced audio is reproduced. It is an object of the present invention to provide a compressed data decoding apparatus and method having a sound quality control function that prevents a signal output from being interrupted and is free from noise and does not impair the quality of decoded speech.

[0015]

In order to achieve the above object, the present invention is configured as follows. That is, [1] First, a data processing unit that receives a data stream of content that has been compression-encoded and performs a decoding process according to a quality parameter, and stores decoded data that has been decoded by the data processing unit. In a compressed data decoding apparatus having a buffer and analog signal conversion means for reading decoded data stored in the buffer at a predetermined sampling rate and performing analog signal processing according to a quality parameter, the decoding processing and the decoding processing are performed according to the state of the apparatus. The apparatus is characterized in that it has a changing means for changing the quality parameter for at least one of the analog signal processing.

In decoding the compressed and encoded content data, the compressed data decoding device receives the stream of the content data, decodes the stream, and stores the processed data in a buffer. In the present invention, a quality parameter is determined, data processing is performed in data processing according to the quality parameter, and analog signal conversion is performed by analog signal conversion means. Quality (function by changing means for changing the quality parameter).

In the present invention, it is assumed that the quality parameter depends on, for example, the state of the compressed data decoding device or the data processing state for decoding the stream. That is, the quality parameters of the decoding process and the reproduction process are varied according to the situation, and the decoding process speed and the D
Adjust the / A conversion speed. This makes it possible to adjust the processing load, and thus, if the content data is, for example, audio data, it is possible to prevent the occurrence of “interruption” in the audio output that causes a large noise in the audio. Thus, the quality of the audio signal output can be maintained.

[2] According to a second aspect of the present invention, in the compressed data decoding apparatus according to the above item [1], when there is a speed difference between the decoding process and the analog signal conversion process, the speed difference is reduced. To this end, there is provided a changing means for changing at least one of the quality parameters of the decoding process and the analog signal conversion process.

The changing means (quality parameter determining means) for changing the quality parameter compares the decoding processing speed (decoding processing speed) of the data processing means with the sampling speed of the analog converting means, and determines the quality parameter according to the comparison result. I do.

According to this configuration, the quality parameter compares the decoding processing speed (decoding processing speed) with the sampling speed of the analog signal conversion (D / A conversion). The decoding processing load is adjusted by changing the quality of at least one of decoding and reproduction in accordance with the determined quality parameter.
Therefore, the decoding processing speed and D
By changing the sampling speed of at least one of the / A conversion and maintaining the balance between the two speeds, the data in the buffer can always be maintained at a constant amount, and therefore, the reproduction output of high-quality content without "interruption" can be achieved. Will be able to gain.

[3] The third aspect of the present invention is that, in the compressed data decoding device according to the above item [1], if the compressed data decoding device has a power saving function, the quality is determined according to the power saving setting state. It is characterized in that parameters are determined, and the quality of data processing in the data processing means and the conversion quality in the analog signal conversion means are changed according to the determined quality parameters.

When the compressed data decoding device has a power saving function, a quality parameter is determined in accordance with the setting state of the power saving, and the decoding processing load is changed by changing the quality of decoding and reproduction in accordance with the determined quality parameter. To adjust. This prevents "interruption" of the reproduced content. Therefore, in a device having a function that affects the operation speed of the device, such as a power saving setting, a parameter that determines the quality of decoding and playback is determined according to the power saving setting of the device, and decoding and playback are performed according to the parameter. In this case, the load of the decoding process can be adjusted by changing the quality of the content, so that, for example, if the content data is audio data, in this case, it is possible to prevent the occurrence of "interruption" of the sound. To enable high-quality audio output.

[4] According to the present invention, first, in the compressed data decoding apparatus according to the above item [1], when the battery is used as a main power supply, the quality parameter determining means determines the quality parameter according to the consumption state of the battery. Is determined.

According to the present invention, the power saving effect is enhanced by reducing the decoding processing load according to the consumption state of the battery. When the compressed data decoding device uses a battery as a main power source, the compressed data decoding device determines a quality parameter. Corresponds to the consumption state of the battery. However, the quality parameter is determined in accordance with the consumption state of the battery, and the decoding process is performed by changing the quality of decoding and reproduction in accordance with the quality parameter. Since the load can be adjusted, the operation time during which the content can be reproduced by the battery can be extended.

[5] Fifth, according to the present invention, upon decoding the data of the compression-encoded content, the stream of the content data is received and decoded by decoding means, and inversely quantized by the inverse quantization means. In a compressed data decoding apparatus which quantizes and restores audio data, stores the data in a buffer, reads the data stored in the buffer at a predetermined sampling rate by an analog signal conversion means, converts the data into an analog signal, and outputs the analog signal A parameter determining means for determining a parameter in accordance with a state of the compressed data decoding apparatus (for example, a power supply state) or a data processing state for decoding a stream. The bit width of the coded signal is varied according to the parameters determined by the determining means. The features.

According to the present invention, the inverse quantization means (inverse MDCT
Unit), a means for varying the decoding processing speed according to the data processing status for decoding the stream or according to the state of the compressed data decoding device (for example, using a battery (battery) as the main power supply) In such a case, the mode is determined according to the remaining capacity of the battery (remaining battery capacity), or when the mode of the regeneration is configured to be switchable between normal and energy saving modes. (Depending on the state of the device), the parameters are determined, and the data bit width in decoding the encoded signal is changed corresponding to the parameters,
When there is a danger that the processing will be slowed down, the number of calculations is reduced by performing a calculation process using a signal with a small number of digits (reducing the number of digits of data), and the decoding processing speed is improved.
This makes it possible to prevent interruption of content reproduction.

[6] Sixth, according to the present invention, when decoding the data of the content which has been compression-encoded, the stream of the content data is received and decoded by the decoding means, and inversely quantized by the inverse quantization means. In a compressed data decoding device which quantizes and restores the data, accumulates the data in a buffer, and reads out the data accumulated in the buffer at a predetermined sampling rate by an analog signal conversion means to convert it into an analog signal and outputs the analog signal, The state of the compressed data decoding device (for example, the state of the power supply), or
According to a data processing situation for decoding the stream, a parameter determining means for determining a parameter is provided,
The decoding means is characterized in that the bit width of data when decoding the coded signal is varied in accordance with the parameters determined by the parameter determining means.

According to the present invention, the means for varying the decoding processing speed depends on the data processing status for decoding the stream or according to the state of the compressed data decoding device (for example, a battery (battery) is connected to the main power supply). In the case where the battery is used, the mode is determined according to the remaining capacity of the battery (remaining battery capacity) or in the case where the regeneration is configured to be switchable between normal and energy saving modes. In other words, the parameters are determined (according to the state of the apparatus), and the upper limit of the decoding frequency for decoding the encoded signal is adjusted in accordance with the parameters so that the processing is not slowed down.

That is, in the decoding means (Huffman decoding unit) of the compressed data decoding apparatus, as a method of varying the decoding processing speed, the processing speed is adjusted by adjusting the upper limit of the decoding frequency corresponding to the parameter determined by the parameter determining means. In order to reduce the decoding processing load, the decoding processing load is reduced by determining the upper limit of the band for performing the decoding process on the coded signal expanded in each band, and limiting the band for performing the processing. Improve.

[7] In the seventh aspect of the present invention, upon decoding the compression-encoded content data, the content data stream is received and subjected to data processing for decoding by a data processing means. The processed data is stored in a buffer, read out at a predetermined sampling rate, converted into an analog signal by an analog signal conversion means, and output. The state of the compressed data decoding apparatus (for example, , Power supply status, etc.) or a parameter determining means for determining a parameter in accordance with a data processing state for decoding the stream, wherein the decoding means performs the analog signal conversion in accordance with the parameter determined by the parameter determining means. The sampling speed of the means is changed.

According to the present invention, as means for varying the decoding processing speed, depending on the data processing situation for decoding the stream or according to the state of the compressed data decoding device (for example, a battery (battery) is In the case where the battery is used, the mode is determined according to the remaining capacity of the battery (remaining battery capacity) or in the case where the regeneration is configured to be switchable between normal and energy saving modes. In other words, the parameters are determined (according to the state of the apparatus), and the upper limit of the decoding frequency for decoding the encoded signal is adjusted in accordance with the parameters so that the processing is not slowed down.

Accordingly, even if there is a danger that the data processing for decoding by the data processing means cannot keep up with the analog signal conversion speed by the analog signal conversion means, the analog signal conversion corresponding to the determined parameter by the parameter determination means may occur. By adjusting the sampling speed of the means, it is possible to maintain the balance with the decoding processing speed and to always keep the stock of the decoded data in the data buffer at a constant amount. become.

[0033]

Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) <Structure of First Embodiment> FIG. 1 is a block diagram showing an example of the structure of the apparatus of the present invention. In the figure, reference numeral 101 denotes a storage medium holding compressed audio data, which is a memory card or a magneto-optical disk (MD, MO, CD-ROM, CD-ROM).
R, CD-RW, DVD, etc.), magnetic storage media (FD,
HDD) can be considered.

The compressed audio decoding apparatus with an audio quality function according to the present invention is battery-operated, receives compressed audio data supplied from the storage medium 101, and decodes it into the original data. , Huffman decoding unit 103, inverse quantization unit 104, aliasing reduction butterfly unit 105, inverse MDCT unit 106, subband synthesis filter bank 107, data buffer 108, A /
It comprises the components of a D conversion unit 109, a speed calculation unit 110, and a quality parameter creation unit 111.

Of these, the bit stream developing unit 1
02 develops the input bit stream into side information (scale factor, decoding table, etc.) and content (audio data).

The Huffman decoding unit 103 converts the content expanded by the bit stream expansion unit 102 into an encoded signal of each band.
Reference numeral 04 denotes a process for performing inverse quantization on the coded signal of each band converted by the bit stream expansion unit 102 and returning the coded signal to the original Huffman code.

Further, the folding distortion removing butterfly 105 is used.
Is to remove aliasing distortion from the coded signal (Huffman coded signal) after inverse quantization by the Huffman decoding unit 103.

The inverse MDCT section 106 has a Huffman decoding section 1
A signal (encoded signal) that has been returned to the original Huffman code at 03 and has been subjected to aliasing removal by the aliasing removal butterfly 105 is converted into a distribution on the signal frequency axis, and is distributed on the time axis. Is converted into a distribution on the signal frequency axis.

The sub-band synthesis filter bank 10
Reference numeral 7 denotes a unit for performing a process of combining encoded signals converted into a distribution on the time axis by the inverse MDCT unit 106, and combines encoded signals of respective bands into one decoded data. .

The data buffer 108 is for temporarily holding the decoded data synthesized by the sub-band synthesis filter bank 107, and the D / A converter 109
Is for extracting (sampling) the decoded data stored in the data buffer 108 at a fixed timing, converting the decoded data into analog audio (analog signal), and outputting the analog audio to the outside.

The speed calculation unit 110 determines the decoding processing speed by measuring the time required from the input of the stream to the generation of the decoded data, and compares it with the sampling speed to determine the speed difference, or The speed difference is obtained from the timing of increase or decrease of the free space in the data buffer 108. Specifically, the development speed of the stream developed by the bit stream development unit 102 is obtained as information, and the subband synthesis filter bank 107 The amount of decoded data to be synthesized and output is obtained as information, the time required from the input of the stream to the generation of decoded data is measured, the decoding processing speed is determined, and this and the A / D conversion unit are determined. 109 to determine the speed difference by comparing with the sampling speed, or Determining the speed difference from the timing of the increase or decrease of the free space of Tabaffa 108.

The quality parameter creation unit 111 creates a quality parameter corresponding to the speed difference based on the information on the speed difference obtained by the speed calculation unit 110. Here, the quality parameter indicates the data length per sample when the decoding processing speed is equal to or higher than the sampling speed in the A / D converter 109, for example, 3
If the decoding processing speed is 2 [bit] (high quality, that is, high resolution) and the decoding processing speed cannot catch up with the sampling speed in the A / D conversion unit 109, the configuration data length per sample is set to, for example, 16 [bits]. bit] (low quality, that is, low resolution), an appropriate quality parameter of several stages is set in advance, and an optimal quality parameter is generated at that time corresponding to the speed difference.

Accordingly, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly unit 105, the inverse M
The DCT section 106 and the sub-band synthesis filter bank 107 are mode switching type processing systems that can perform data processing at resolutions corresponding to the respective modes so that data processing can be performed in quality modes corresponding to the respective quality parameters. Further, the A / D converter 109 is configured to convert digital data into an analog signal in a data bit configuration corresponding to a quality parameter. By doing so, it is possible to process the entire system corresponding to the quality parameter.

That is, according to the value of the quality parameter created by the quality parameter creation unit 111, the decoding processing speed and the sampling speed are varied to balance the two speeds, and the stock of the decoded data in the data buffer 108 is maintained at a constant amount. In this configuration, the entire system can be controlled so that a situation in which decoded data is insufficient during sampling does not occur.

<Explanation of Function of First Embodiment> Next,
The operation of the present device having the above configuration will be described. First, a flow of a general audio decoding process will be described.

It is assumed that contents (audio data, etc.) are compressed and encoded and stored together with side information (scale factors, decoding tables, etc.) in the storage medium 101.

In this state, when an external reproduction request is made, a bit stream to be reproduced is output from the storage medium 101. Bit stream expansion unit 102
Expands the input bit stream into side information (scale factor, decoding table, etc.) and content (audio data).

The Huffman decoding unit 103 shown in FIG. 1 converts the content expanded by the bit stream expansion unit 102 into an encoded signal of each band, and supplies it to the inverse quantization unit 104. Then, the inverse quantization unit 104 performs inverse quantization on the coded signal of each band provided by the conversion.

On the other hand, the folding distortion removing butterfly 105
Performs aliasing removal on the coded signal after the inverse quantization. Then, the coded signal from which the aliasing distortion has been removed is provided to inverse MDCT section 106. Inverse MDCT unit 10
In step 6, the coded signal distributed on the time axis is converted into a distribution on the frequency axis, and this is supplied to the sub-band synthesis filter bank 107. Subband synthesis filter bank 1
At 07, the encoded signals of the respective bands are combined to form one piece of decoded data, which is provided to the data buffer 108.

The data buffer 108 sequentially stores the decoded data synthesized by the sub-band synthesis filter bank 107.

The D / A conversion unit 109 includes a data buffer 1
08 at a fixed timing, and retrieves (samples) the decoded data from the oldest data in the storage order,
Convert to analog audio and output to outside.

Here, the bit stream developing unit 102,
The processing of the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing distortion reduction butterfly 105, the inverse MDCT unit 106, and the subband synthesis filter bank 107 is the speed in the normal mode, and the data resolution is compatible with 32 [bit]. is there.

Here, it is assumed that an input stream having a large decoding processing load is output from the storage medium 101.
In such a case, the decoding processing speed is reduced, and the speed at which the decoded data is stored in the data buffer 108 is lower than the sampling speed of the D / A converter 109.

If this state continues, the supply-demand relationship of the decrypted data is broken, and the supply of the decrypted data cannot keep up.
Until the decoded data stored in the data buffer 108 runs short, the D / A conversion unit 109 outputs “silence” at each sampling until the next decoding process is completed.
Will be output. This “silence” period is a “break” in the sound.

In the present invention, a monitoring control system is provided to prevent the occurrence of the "silence" state. That is, in the speed calculation unit 110, which is one component of the monitoring control system, the decoding processing speed is obtained from the change in the free space in the data buffer 108 or the speed at which a fixed amount of decoded data is generated. Is calculated. Then, this is notified to the quality parameter creation unit 111.

Then, in the quality parameter creating section 111 which is one component of the monitoring control system, the speed calculating section 110
A quality parameter is created according to the speed difference information notified from. Then, the decoding processing speed or the sampling speed is changed using the created quality parameter, so that the stock of the decoded data in the data buffer 108 is adjusted to be always a constant amount. In this way, by preventing a situation in which the decoded data is insufficient, "interruption" of audio is prevented.

That is, in this case, the quality parameter generating section 111 generates a quality parameter for the low quality, which is output to the Huffman decoding section 103 and the inverse quantization section 1.
04, aliasing reduction butterfly 105, inverse MDCT
To the D / A conversion unit 109.

As a result, the Huffman decoding unit 10
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
9 is the half of the processing corresponding to the resolution of 32 [bit]
The mode is changed to the processing corresponding to the resolution of 6 [bit]. Therefore, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly 105, and the inverse MDCT unit 1
In step 06, the processing becomes lighter and the decoding processing becomes smoother.
Then, the decoded audio data is sequentially stored in the data buffer 108.

On the other hand, the D / A converter 109 reads out audio data from the data buffer 108 at each normal sampling timing, converts the audio data into an analog signal, and outputs it as an audio signal. At this time, the audio data from the data buffer 108 is data having a resolution of 16 [bit].
Since the low-quality data is received as a parameter from the quality parameter generation unit 111, the D / A conversion unit 10
9 performs processing with a resolution of 16 [bits], and although the sound quality is somewhat reduced, the processing speed increases as the processing becomes lighter, and the situation of insufficient decoded data in the data buffer 108 does not occur. .

As a result, the audio signal is not interrupted, and as a result, the quality can be greatly improved as compared with the related art.

Since the sampling timing of the D / A conversion unit 109 is constant, if the data processing continues to be light, the D / A conversion cannot catch up, and the data buffer 10
Voice data accumulates in 8.

When such a situation occurs, the following operation is performed. That is, in the speed calculation unit 110, which is one component of the monitoring control system, the decoding processing speed is obtained from the change in the free space in the data buffer 108 or the speed at which a fixed amount of decoded data is generated. Is calculated. Then, this is notified to the quality parameter creation unit 111.

Therefore, in the quality parameter creation unit 111 which is one component of the monitoring control system, the speed calculation unit 110
A quality parameter is created according to the speed difference information notified from. Then, the decoding processing speed or the sampling speed is changed using the created quality parameter, so that the stock of the decoded data in the data buffer 108 is adjusted to be always a constant amount. In this way, by preventing a situation in which the decoded data is insufficient, it is possible to prevent the audio data from being accumulated more than necessary in the data buffer 108, and the resolution of the data is not always kept low. To do.

That is, in this case, the quality parameter generating unit 111 generates a quality parameter for the quality parameter, and the quality parameter is generated by the Huffman decoding unit 103 and the inverse quantization unit 1.
04, aliasing reduction butterfly 105, inverse MDCT
To the D / A conversion unit 109.

As a result, the Huffman decoding unit 10
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
9 is 32 times that of processing corresponding to 16 [bit] resolution.
The mode is changed to the processing corresponding to the resolution of [bit]. Therefore, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly 105, and the inverse MDCT unit 1
In step 06, the processing proceeds with the normal resolution of 32 [bit]. Therefore, the quality of the decoding process returns to normal high quality. Then, the decoded audio data is sequentially stored in the data buffer 108.

On the other hand, the D / A converter 109 reads out audio data from the data buffer 108 at each normal sampling timing, converts the data into an analog signal, and outputs the analog signal. At this time, the audio data from the data buffer 108 is 32 bits from data having a resolution of 16 [bit].
Although the data has returned to the data of [bit] resolution, the D / A conversion unit 109 receives 32 [bi] since the high-quality data is received as a parameter from the quality parameter creation unit 111.
t], and the sound quality returns to the original high quality.

As described above, when the data processing cannot be performed in time, the resolution is reduced and the processing is lightened. If there is a margin in the data processing, the resolution is returned to the original state and the processing is returned to the high-quality data processing. The audio signal is not interrupted due to being unable to make it in time, and when there is no longer a concern that data processing will not be in time, high quality data processing will be returned. Become.

The above is to prevent the interruption of the generated audio signal (audio signal) due to the fact that the data processing cannot be performed in time due to the weight of the input stream when the processing is heavy. , In such situations, the resolution is reduced and the processing is lighter,
In this way, the data processing speed is increased,
If there is room for data processing, the resolution was restored and the processing was returned to high-quality data processing.However, in the case of a device with a power saving mode, when the operation was in the power saving mode In addition, the decoding processing speed is slowed down, and the stock of decoded data in the data buffer 108 becomes insufficient, so that "breakage" of audio occurs. An embodiment capable of preventing this will be described below as a second embodiment.

(Second Embodiment) <Structure of Second Embodiment> FIG. 2 is a block diagram showing an example of the structure of a battery-driven type device of the present invention in which a power saving mode can be used. In the figure, reference numeral 101 denotes a storage medium holding compressed audio data, which is a memory card or a magneto-optical disk (MD, MO, CD-ROM, CD-ROM).
R, CD-RW, DVD, etc.), magnetic storage media (FD,
HDD) can be considered.

A compressed audio decoding apparatus with an audio quality function according to the present invention is battery-driven, receives compressed audio data supplied from a storage medium 101, and decodes it into original data. , Huffman decoding unit 103, inverse quantization unit 104, aliasing reduction butterfly unit 105, inverse MDCT unit 106, subband synthesis filter bank 107, data buffer 108, A /
D conversion unit 109, speed calculation unit 110, quality parameter creation unit 111, power saving setting unit 112, power saving control unit 11
3, the battery 114, and the clock generator 115.

Of these, the battery 114 is the main power supply of the compressed speech decoding apparatus with the speech quality function according to the present invention.
Each part is operated using the energy of the battery 114. Also, a bit stream developing unit 1
02 develops the input bit stream into side information (scale factor, decoding table, etc.) and content (audio data).

The Huffman decoding unit 103 converts the content expanded by the bit stream expansion unit 102 into an encoded signal of each band.
Reference numeral 04 denotes a process for performing inverse quantization on the coded signal of each band converted by the bit stream expansion unit 102 and returning the coded signal to the original Huffman code.

Also, the folding distortion removing butterfly 105
Is to remove aliasing distortion from the coded signal (Huffman coded signal) after inverse quantization by the Huffman decoding unit 103.

The inverse MDCT section 106 has the Huffman decoding section 1
03, a signal (encoded signal) that has been returned to the original Huffman code and has been subjected to aliasing removal by the aliasing removal butterfly 105 is converted into a distribution on the time axis, and is distributed on the signal frequency axis. Is converted into a distribution on the time axis.

The sub-band synthesis filter bank 10
Reference numeral 7 denotes a unit for performing a process of combining encoded signals converted into a distribution on the time axis by the inverse MDCT unit 106, and combines encoded signals of respective bands into one decoded data. .

The data buffer 108 is for temporarily holding the decoded data synthesized by the sub-band synthesis filter bank 107 and includes a D / A converter 109.
Is for extracting (sampling) the decoded data stored in the data buffer 108 at a fixed timing, converting the decoded data into analog audio (analog signal), and outputting the analog audio to the outside.

The speed calculation unit 110 determines the decoding processing speed by measuring the time required from the input of the stream to the generation of the decoded data, compares it with the sampling speed, obtains the speed difference, or The speed difference is obtained from the timing of increase or decrease of the free space in the data buffer 108. Specifically, the development speed of the stream developed by the bit stream development unit 102 is obtained as information, and the subband synthesis filter bank 107 The amount of decoded data to be synthesized and output is obtained as information, the time required from the input of the stream to the generation of decoded data is measured, the decoding processing speed is determined, and this and the A / D conversion unit are determined. 109 to determine the speed difference by comparing with the sampling speed, or Determining the speed difference from the timing of the increase or decrease of the free space of Tabaffa 108.

The quality parameter creating section 111 creates a quality parameter corresponding to the speed difference based on the information on the speed difference obtained by the speed calculating section 110. Here, the quality parameter indicates the data length per sample when the decoding processing speed is equal to or higher than the sampling speed in the A / D converter 109, for example, 3
If the decoding processing speed is 2 [bit] (high quality, that is, high resolution) and the decoding processing speed cannot catch up with the sampling speed in the A / D conversion unit 109, the configuration data length per sample is set to, for example, 16 [bits]. bit] (low quality, that is, low resolution), an appropriate quality parameter of several stages is set in advance, and an optimal quality parameter is generated at that time corresponding to the speed difference.

Accordingly, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly unit 105, the inverse M
The DCT section 106 and the sub-band synthesis filter bank 107 are mode switching type processing systems that can perform data processing at resolutions corresponding to the respective modes so that data processing can be performed in quality modes corresponding to the respective quality parameters.

The Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly unit 105, the inverse MD
CT section 106 and subband synthesis filter bank 1
Reference numeral 07 denotes an audio decoding processing unit realized by software using the CPU 100.

The A / D converter 109 is configured to convert digital data into an analog signal in a data bit configuration corresponding to a quality parameter. By doing this,
It is possible to process the entire system according to the quality parameters.

That is, according to the value of the quality parameter created by the quality parameter creating unit 111, the decoding processing speed and the sampling speed are varied to balance the two speeds, and the stock of the decoded data in the data buffer 108 is kept at a constant amount. In this configuration, the entire system can be controlled so that a situation in which decoded data is insufficient during sampling does not occur.

The power saving setting section 112 is for setting whether to set or cancel the power saving mode, for example. The power saving control section 113 has a function of monitoring the remaining capacity of the battery 114. And when the power saving mode is set by the power saving setting unit 112, when the remaining capacity of the battery falls below a predetermined level, a control command is issued so as to operate in the power saving mode, As a result, the rate of the operation clock signal CK given to the CPU 100 and its peripheral circuits, which are important components of the present apparatus which plays a central role in various controls and arithmetic processing, is reduced by half from, for example, 100 [MHz] in the normal mode. The clock generation unit 115 is controlled so as to drop to 50 [MHz].

For example, when the operating speed of the CPU 100 is changed (for example, the clock speed is reduced), the operating speed of the entire apparatus is reduced, and the decoding processing speed (computation processing, etc.) is accordingly reduced. This is an operation mode utilizing the fact that the power can be significantly reduced.

The clock generation unit 115 sets the rate of the generated clock signal CK to, for example, 100 [MHz] in the normal mode, or 5 which is half that in the power saving mode.
The switching can be controlled so as to drop to 0 [MHz].

<Operation of the Second Embodiment> Next, the operation of the present apparatus having the above-described configuration will be described. First, a flow of a general audio decoding process will be described.

It is assumed that contents (audio data, etc.) are compressed and encoded in the storage medium 101 together with side information (scale factors, decoding tables, etc.).

In this state, when a reproduction request is made from outside, a bit stream to be reproduced is output from the storage medium 101. Bit stream expansion unit 102
Expands the input bit stream into side information (scale factor, decoding table, etc.) and content (audio data).

The Huffman decoding unit 103 converts the content expanded by the bit stream expanding unit 102 into an encoded signal of each band, and supplies it to the inverse quantization unit 104. Then, the inverse quantization unit 104 performs inverse quantization on the coded signal of each band provided by the conversion.

On the other hand, the folding distortion removing butterfly 105
Performs aliasing removal on the coded signal after the inverse quantization. Then, the coded signal from which the aliasing distortion has been removed is provided to inverse MDCT section 106. Inverse MDCT unit 10
In step 6, the coded signal distributed on the time axis is converted into a distribution on the frequency axis, and this is supplied to the sub-band synthesis filter bank 107. Subband synthesis filter bank 1
At 07, the encoded signals of the respective bands are combined to form one piece of decoded data, which is provided to the data buffer 108.

The data buffer 108 sequentially stores the decoded data synthesized by the sub-band synthesis filter bank 107.

The D / A conversion unit 109 includes a data buffer 1
08 at a fixed timing, and retrieves (samples) the decoded data from the oldest data in the storage order,
Convert to analog audio and output to outside.

Here, the bit stream developing unit 102
The processing of the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing distortion reduction butterfly 105, the inverse MDCT unit 106, and the subband synthesis filter bank 107 is the speed in the normal mode, and the data resolution is compatible with 32 [bit]. is there.

In the present apparatus shown in FIG.
There is a function 2 for canceling the setting of the power saving mode. The setting of such a function is manually performed, for example. When the power saving mode is set, the power saving control unit 113 checks the battery remaining capacity by its own battery remaining capacity monitoring function, and executes the power saving control when the remaining capacity is smaller than a predetermined reference value. . When the power saving mode is cancelled, the normal power saving operation is not performed even if the state of charge is small.

When the power saving mode is set and it is detected that the remaining battery capacity is low, the power saving control unit 1
13 shifts the compressed speech decoding device to the operation in the power saving mode.

This is because, for example, the operation speed of the CPU 100 provided as a center of control of the compressed speech decoding apparatus is as follows.
The power consumption is reduced by lowering the frequency of the clock signal CK, for example.

More specifically, the clock generator 115 is of a variable clock type or a switchable type, and the power saving controller 113 is configured to change the clock frequency in accordance with the setting of the power saving setting unit 112. Part 115
Control.

Then, the peripheral circuit of the CPU is also operated by supplying the clock signal CK having the reduced clock frequency in order to operate in accordance with the clock frequency, so that the operation speed of the entire device is reduced, Accompanying this, the decoding processing speed (calculation processing etc.) becomes slow.

Assuming that the sampling speed in the D / A converter 109 is constant, as described above, if the decoding processing speed (decoded data generation speed) is lower than the sampling speed in the D / A converter 109, the data buffer The audio data stored in the memory 108 cannot be replenished in time, so that the audio data becomes empty and "interruption" occurs in the audio.

In such a case, the quality parameter is created according to the power saving setting, and the decoding processing speed and the sampling speed are adjusted according to the quality parameter, so that the stock of the decoded data in the data buffer 108 is maintained at a constant amount. In addition, the situation where the decoded data is insufficient at the time of sampling is eliminated, thereby preventing the "interruption" of the sound.

Specifically, upon receiving a control signal from power saving control section 113, quality parameter creating section 111 creates a quality parameter according to the power saving setting. In this case, as the quality parameter, the quality parameter creation unit 111
A low-quality one is generated, and the low-quality one is generated.
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
Give 9

As a result, the Huffman decoding unit 10
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
9 is the half of the processing corresponding to the resolution of 32 [bit]
The mode is changed to the processing corresponding to the resolution of 6 [bit]. Therefore, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly 105, and the inverse MDCT unit 1
In step 06, the processing becomes lighter and the decoding processing becomes smoother.
Then, the decoded audio data is sequentially stored in the data buffer 108.

On the other hand, the D / A converter 109 reads out audio data from the data buffer 108 at each normal sampling timing, converts the audio data into an analog signal, and outputs it as an audio signal. At this time, the audio data from the data buffer 108 is data having a resolution of 16 [bit].
Since the low-quality data is received as a parameter from the quality parameter generation unit 111, the D / A conversion unit 10
9 performs processing with a resolution of 16 [bits], and although the sound quality is somewhat reduced, the processing speed increases as the processing becomes lighter, and the situation of insufficient decoded data in the data buffer 108 does not occur. .

Thus, in the power saving mode, the CP
Although the operation clocks of the peripheral circuits of the CPU, such as U100, become slow in response to the power saving mode, the resolution of the audio data is reduced to reduce the processing so that the data processing cannot be completed in time. As a result, the audio signal does not “interrupt” due to the lack of time, so that the quality can be greatly improved as compared with the related art.

Here, it is assumed that battery 114 has been charged or replaced and its remaining capacity has become sufficient. In this case, the power saving control unit 113 checks that the remaining battery capacity is sufficient by its own remaining battery capacity monitoring function, detects that the remaining battery capacity is sufficient, and cancels the power saving control.

As a result, clock signal CK of normal 100 [MHz] is generated from clock generation section 115,
It is supplied to the CPU 100 and CPU peripheral circuits.

In this state, when an external reproduction request is made, a bit stream to be reproduced is output from the storage medium 101. Bit stream expansion unit 102
Expands the input bit stream into side information (scale factor, decoding table, etc.) and content (audio data).

The Huffman decoding unit 103 converts the content expanded by the bit stream expansion unit 102 into an encoded signal of each band, and supplies it to the inverse quantization unit 104.
Then, the inverse quantization unit 104 performs inverse quantization on the coded signal of each band provided by the conversion.

On the other hand, the folding distortion removing butterfly 105
Performs aliasing removal on the coded signal after the inverse quantization. Then, the coded signal from which the aliasing distortion has been removed is provided to inverse MDCT section 106. Inverse MDCT unit 10
In step 6, the coded signal distributed on the time axis is converted into a distribution on the frequency axis, and this is supplied to the sub-band synthesis filter bank 107. Subband synthesis filter bank 1
At 07, the encoded signals of the respective bands are combined to form one piece of decoded data, which is provided to the data buffer 108.

The data buffer 108 sequentially stores the decoded data synthesized by the sub-band synthesis filter bank 107.

The D / A conversion unit 109 is provided with the data buffer 1
08 at a fixed timing, and retrieves (samples) the decoded data from the oldest data in the storage order,
Convert to analog audio and output to outside.

Here, the bit stream developing unit 102
The processing of the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing distortion reduction butterfly 105, the inverse MDCT unit 106, and the subband synthesis filter bank 107 is the speed in the normal mode, and the data resolution is compatible with 32 [bit]. is there.

[0114] When transitioning in such a state, an input stream having a large decoding processing load is stored in the storage medium 1.
It is assumed that the data is output from 01. In such a case, the decoding processing speed is reduced, and the speed at which the decoded data is stored in the data buffer 108 is lower than the sampling speed of the D / A converter 109.

If this state continues, the supply and demand relationship of the decrypted data will collapse, and the supply of the decrypted data will not be able to keep up.
Until the decoded data stored in the data buffer 108 runs short, the D / A conversion unit 109 outputs “silence” at each sampling until the next decoding process is completed.
Will be output. This “silence” period is a “break” in the sound.

In the present invention, a monitoring control system is provided to prevent the occurrence of the "silence" state. That is, in the speed calculation unit 110, which is one component of the monitoring control system, the decoding processing speed is obtained from the change in the free space in the data buffer 108 or the speed at which a fixed amount of decoded data is generated. Is calculated. Then, this is notified to the quality parameter creation unit 111.

Then, in the quality parameter creating section 111 which is one component of the monitoring control system, the speed calculating section 110
A quality parameter is created according to the speed difference information notified from. Then, the decoding processing speed or the sampling speed is changed using the created quality parameter, so that the stock of the decoded data in the data buffer 108 is adjusted to be always a constant amount. In this way, by preventing a situation in which the decoded data is insufficient, "interruption" of audio is prevented.

That is, in this case, the quality parameter generating unit 111 generates a quality parameter for the low quality, and this is generated by the Huffman decoding unit 103 and the inverse quantization unit 1.
04, aliasing reduction butterfly 105, inverse MDCT
To the D / A conversion unit 109.

As a result, the Huffman decoding unit 10
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
9 is the half of the processing corresponding to the resolution of 32 [bit]
The mode is changed to the processing corresponding to the resolution of 6 [bit]. Therefore, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly 105, and the inverse MDCT unit 1
In step 06, the processing becomes lighter and the decoding processing becomes smoother.
Then, the decoded audio data is sequentially stored in the data buffer 108.

On the other hand, the D / A converter 109 reads out audio data from the data buffer 108 at each normal sampling timing, converts the audio data into an analog signal, and outputs it as an audio signal. At this time, the audio data from the data buffer 108 is data having a resolution of 16 [bit].
Since the low-quality data is received as a parameter from the quality parameter generation unit 111, the D / A conversion unit 10
9 performs processing with a resolution of 16 [bits], and although the sound quality is somewhat reduced, the processing speed increases as the processing becomes lighter, and the situation of insufficient decoded data in the data buffer 108 does not occur. .

As a result, the audio signal is not interrupted, and as a result, the quality can be greatly improved as compared with the related art.

Since the sampling timing of the D / A conversion unit 109 is constant, if the state of light data processing continues, the D / A conversion cannot catch up and the data buffer 10
Voice data accumulates in 8.

In such a situation, the present apparatus operates as follows. That is, in the speed calculation unit 110 which is one component of the monitoring control system, the data buffer 1
The decoding processing speed is obtained from the change in the free space in 08 or the speed at which a fixed amount of decoded data is generated, and the speed difference from the current sampling speed is calculated. And
This is notified to the quality parameter creation unit 111.

Therefore, in the quality parameter creation unit 111 which is one component of the monitoring control system, the speed calculation unit 110
A quality parameter is created according to the speed difference information notified from. Then, the decoding processing speed or the sampling speed is changed using the created quality parameter, so that the stock of the decoded data in the data buffer 108 is adjusted to be always a constant amount. In this way, by preventing a situation in which the decoded data is insufficient, it is possible to prevent the audio data from being accumulated more than necessary in the data buffer 108, and the resolution of the data is not always kept low. To do.

That is, in this case, the quality parameter generating unit 111 generates a quality parameter for the quality parameter, and this is output to the Huffman decoding unit 103 and the inverse quantization unit 1.
04, aliasing reduction butterfly 105, inverse MDCT
To the D / A conversion unit 109.

Thus, the Huffman decoding unit 10
3. Inverse quantization section 104, aliasing reduction butterfly 1
05, inverse MDCT section 106, and D / A conversion section 10
9 is 32 times that of processing corresponding to 16 [bit] resolution.
The mode is changed to the processing corresponding to the resolution of [bit]. Therefore, the Huffman decoding unit 103, the inverse quantization unit 104, the aliasing reduction butterfly 105, and the inverse MDCT unit 1
In step 06, the processing proceeds with the normal resolution of 32 [bit]. Therefore, the quality of the decoding process returns to normal high quality. Then, the decoded audio data is sequentially stored in the data buffer 108.

On the other hand, the D / A converter 109 reads out the audio data from the data buffer 108 at each normal sampling timing, converts the audio data into an analog signal, and outputs it as an audio signal. At this time, the audio data from the data buffer 108 is 32 bits from data having a resolution of 16 [bit].
Although the data has returned to the data of [bit] resolution, the D / A conversion unit 109 receives 32 [bi] since the high-quality data is received as a parameter from the quality parameter creation unit 111.
t], and the sound quality returns to the original high quality.

That is, in the normal mode, when the data processing cannot be performed in time, the resolution is reduced to reduce the processing, and if there is enough data processing, the resolution is returned to the original state and the processing is returned to the high quality data processing. Thus, the audio signal is not interrupted due to the inability to process the data, and when there is no longer a concern that the data cannot be processed in time, the process returns to high-quality data processing.
The quality can be greatly improved as compared with the conventional case.

According to the configuration of the second embodiment, if data processing cannot be performed in time, the resolution is reduced to reduce the processing, and if there is room in the data processing, the resolution is returned to the original state to perform high-quality data processing. In order to prevent the audio signal from being interrupted due to the inability to perform the data processing in time, and to eliminate the concern that the data processing cannot be performed in time, a quality-oriented function for returning to the high-quality data processing is provided. Mode), and a function having two types of functions corresponding to the power saving mode when the remaining battery capacity is reduced can be obtained. It looks like 3.

First, the current remaining battery level is checked (step S201 in FIG. 3). As a result, if the remaining battery power is below a certain threshold value, the quality parameter is changed so as to perform low-resolution processing (step S in FIG. 3).
206) The decoding processing speed and the sampling speed are changed (step S207 in FIG. 3).

On the other hand, as a result of checking the current remaining battery level, if the remaining battery level is equal to or greater than a certain threshold value, it is checked whether or not the apparatus is operating in the power saving mode (step S202 in FIG. 3). If the operation mode is the power mode, the quality parameter is changed so that the processing is performed at a high resolution (step S206 in FIG. 3), and the decoding processing speed and the sampling speed are changed (step S20 in FIG. 3).
7).

As a result of the confirmation in step S202, if the operation is not in the power saving mode, the speed difference between the decoding processing speed and the sampling speed is obtained (step S203 in FIG. 3). It is determined whether or not it is faster than the decoding processing speed (step S204 in FIG. 3).

As a result, when it is determined that the sampling speed is faster than the decoding processing speed (step S2 in FIG. 3).
04), and a quality parameter (quality parameter for lowering the resolution) corresponding to the speed difference is created (step S2 in FIG. 3).
06), thereby improving the decoding processing speed and suppressing the sampling speed (step S207 in FIG. 3).

If it is determined in step S204 that the decoding processing speed is faster than the sampling speed (step S205 in FIG. 3), a quality parameter (a quality parameter for increasing the resolution) corresponding to the speed difference is created (FIG. 3). Step S208), the decoding processing speed and the sampling speed are adjusted (Step S20 in FIG. 3).
9) Improve sound quality. If the decoding processing speed and the sampling speed are balanced, the processing speed is maintained as it is.

By repeating these operations at all times and maintaining the balance between the decoding processing speed and the sampling speed, the stock of decoded data in the data buffer 108 is stabilized at a constant amount. And this allows us to prevent "breaks" in the sound,
In addition, it is possible to perform reproduction with quality corresponding to the remaining battery capacity.

Next, some examples of techniques for increasing the data processing speed will be described.

(Third Embodiment) An example of a technique for speeding up processing will be described. In the subband synthesis filter bank 107 in the configuration of FIGS. 1 and 2, for example, when performing a synthesis operation using data of a 32 [bit] configuration (ie, a 4-byte configuration) per sample of an encoded signal, Depending on the processing capacity of the CPU, it is necessary to repeat the calculation several times, so that it takes time to process.

To reduce the processing time, it is conceivable to reduce the load of the arithmetic processing.

In order to reduce the load of the arithmetic processing, for example,
Normally, a coded signal having a 32 [bit] configuration per sample is used. In the low-quality mode, for example, the coding signal processing is performed using a coded signal having a 16 [bit] configuration per sample.

That is, in the low quality mode, the calculation accuracy is set to half the accuracy of the high quality mode which is the normal mode.

This corresponds to the sub-band synthesis filter bank 1
07 is a parameter corresponding to the low-quality mode, whereby the sub-band synthesis filter bank 107 converts the data (coded signal) of 32 [bit] per sample, which is the precision corresponding to the high-quality mode, into one.
The accuracy is changed to data (encoded signal) having a configuration of 16 [bits] per sample, and the inverse MDCT unit 106
The data after DCT processing is 16 [bit] per sample
Is converted into a coded signal having the above configuration, and the combining operation is performed using a coded signal having a small number of digits.

As a result, the number of operations can be reduced and the decoding processing speed can be improved because the accuracy is reduced.

(Fourth Embodiment) When a content (audio data) is encoded by Huffman coding technology, the Huffman decoding unit 103 converts the input content (audio data) into each of them according to a decoding table. At this time, a specific band is masked according to the value of the quality parameter created by the quality parameter creation unit 111 from the encoded signal of each band.

At this time, the coded signal of the masked band is supplied from the subsequent inverse quantization section 104 to the D / A conversion section 1.
At the 09th stage, it is no longer used.

Thus, by limiting the band for performing the decoding process, the load can be reduced, and the decoding process speed can be improved.

(Fifth Embodiment) Next, a technique for eliminating interruption due to lack of decoded data will be described. 1 and 2, the D / A converter 109 extracts the decoded data from the data buffer 108 at a constant sampling rate.

To ensure that there is no shortage of decrypted data,
It is possible to cope with this by changing the sampling speed. Specifically, the sampling rate is varied according to the quality parameter created by the quality parameter creation unit 111.

As a result, it is possible to balance the decoding processing speed, maintain the stock of the decoded data in the data buffer 108 at a fixed amount, and prevent a situation that the decoded data is insufficient during sampling from occurring. Therefore, it is possible to prevent the "interruption" of the sound.

(Sixth Embodiment) Next, the speed calculation unit 110
A method for obtaining a speed difference between the decoding processing speed and the sampling speed in the method will be described. One of the methods is the speed calculation unit 11
0, the free space in the data buffer 108 is monitored, and the time difference between the timing at which the decoded data is stored (decoding process completion timing) and the timing at which the decoded data is taken out (sampling timing) is obtained. This is a method of calculating a speed difference between the decoding processing speed and the sampling speed.

(Seventh Embodiment) The speed calculation unit 110
As another method of obtaining the speed difference between the decoding processing speed and the sampling speed in the speed calculation unit 11 shown in FIG.
0, the time from when the stream is input from the storage medium 101 to the 2-bit stream expansion unit 102 until the decoded data is generated by the sub-band synthesis filter bank 107 is measured, and the sampling time and the sampling speed are used as the decoding processing time. And a speed difference is calculated.

Various embodiments have been described above. In short, the present invention performs data processing for decoding when decoding compression-encoded content data, and stores the processed data in a buffer. Then, in a device that converts the signal into an analog signal at a predetermined sampling rate and outputs the signal as an audio signal, the data processing cannot be performed in time, thereby preventing the audio signal from being interrupted due to an empty buffer. In such a situation, in such a situation, reduce the processing by reducing the resolution to speed up the data processing, and if there is room for the data processing, return the resolution to the original and return to the high-quality data processing. And
In the case of a device with a power saving mode, the decoding processing speed is slowed down even when the operation is performed in the power saving mode, and the "interruption" of the audio signal caused by emptying the buffer is reduced, and the resolution is reduced. Therefore, the data processing can be prevented by reducing the data processing speed and speeding up the data processing.

Therefore, when reproducing music content, it is possible to reliably prevent breaks in sound, and to reproduce an audio signal with less noise.

The present invention is not limited to the embodiments described above, but can be implemented with various modifications. For example, in the above-described embodiment, the content converted into the compressed data is audio, but the present invention can be similarly applied to other content other than audio such as moving images and still images.

For example, in the above-described embodiment, two selectable resolutions are used. However, three or more resolutions can be used if necessary. In addition, the clock frequency is set to 100 [MHz] and 50 [MHz] in the embodiment, but this is an example and is not limited to this.

Further, for example, in an apparatus using a battery as a power supply for driving, several levels of quality parameters are created according to the energy consumption state of the battery, and the decoding processing load is reduced according to the crystal quality parameters. Thus, it may be possible to suppress the power consumption in the decoding processing system and perform the sound reproduction with a quality corresponding to the remaining capacity of the battery. The point is, depending on the state of the compressed data decoding device (for example, when a battery (battery) is used as the main power supply, etc., depending on the remaining capacity of the battery (remaining battery capacity), or whether the playback is normal / In the case of a configuration in which the mode can be switched between energy-saving modes, the mode is determined according to the state of the apparatus (e.g., depending on the state of the apparatus), and decoding of the coded signal corresponding to this parameter is performed. By changing the data bit width of, when the processing is slow, the arithmetic processing is performed using a signal with a small number of digits (reducing the number of digits of the data), thereby optimizing the number of calculations according to the situation, "Disruption" of content playback by improving decoding processing speed
It is also possible to suppress the occurrence.

[0156]

As described in detail above, according to the present invention, when decoding data of a compression-encoded content, data processing for decoding is performed, and the processed data is stored in a buffer. However, in a device that reproduces content by converting this to an analog signal at a predetermined sampling rate, it is possible to prevent interruption of content reproduction caused by emptying of a buffer due to a lack of data processing. Therefore, for example, if the content data is audio, it is possible to provide a compressed data signal decoding apparatus capable of reproducing an audio signal with less noise.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the present invention, and is a block diagram showing a first embodiment of the present invention.

FIG. 2 is a diagram for explaining the present invention, and is a block diagram showing a second embodiment of the present invention.

FIG. 3 is a flowchart illustrating a flow of processing according to a second embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 CPU (processor) 101 storage medium 102 bit stream decompression unit 103 Huffman decoding unit 104 inverse quantization unit 105 aliasing reduction butterfly unit 106 inverse MDCT unit 107 subband synthesis filter bank 108 data buffer 109: A / D (digital / analog) converter 110: speed calculator 111: quality parameter generator 112: power saving setting unit 113: power saving controller 114: battery 115: clock generator

Continued on the front page F term (reference) 5D045 DA20 5J064 AA00 AA01 BA09 BB10 BC01 BC02 BC07 BC12 BC14 BD01 9A001 BB01 BB03 BB04 EE05 HH15 HH18 KK31 KK37 KK43 LL02

Claims (13)

[Claims] 1. A data processing means for receiving a data stream of a compression-encoded content and performing a decoding process according to a quality parameter, a buffer for storing decoded data decoded by the data processing means, and a buffer for storing the decoded data. An analog signal converting means for reading out the decoded data stored in the memory at a predetermined sampling rate and performing analog signal processing according to a quality parameter, wherein the decoding processing and the analog signal conversion processing are performed according to the state of the apparatus. Characterized in that the apparatus further comprises changing means for changing the quality parameter for at least one of the compressed data. 2. A data processing means for receiving a compression-encoded data stream of a content and performing a decoding process in accordance with a quality parameter; a buffer for storing decoded data decoded by the data processing means; And an analog signal converting unit that reads out the decoded data stored in the storage unit at a predetermined sampling rate and performs analog signal processing in accordance with a quality parameter, wherein there is a speed difference between the decoding processing and the analog signal conversion processing. In this case, in order to reduce the speed difference, the compressed data decoding apparatus further comprises changing means for changing at least one of the quality parameters of the decoding process and the analog signal conversion process. 3. A data processing means for receiving a data stream of compressed and encoded contents and performing a decoding process in accordance with a quality parameter; a buffer for storing decoded data decoded by the data processing means; An analog signal converting means for reading out the decoded data stored at a predetermined sampling rate and performing analog signal processing in accordance with a quality parameter, when the analog signal processing is slower than the decoding processing, Determining the at least one quality parameter of the decoding process and the analog signal conversion process to eliminate the difference, wherein the data processing unit and the analog signal conversion unit are determined by the determination unit; Characterized by performing processing according to quality parameters Compressed data decoding device. 4. A data processing means for receiving a data stream of compression-encoded content and performing a decoding process in accordance with a quality parameter; a buffer for storing decoded data decoded by the data processing means; Setting means for setting a power saving mode in a compressed data decoding device having an analog signal converting means for reading out decoded data stored in the memory at a predetermined sampling rate and performing analog signal processing according to a quality parameter; If the power mode is set,
Changing means for changing the quality parameter so that at least one of the decoding processing and the analog signal conversion processing is in an operation state with low power consumption. 5. A data processing means for receiving a data stream of compression-encoded content and performing a decoding process in accordance with a quality parameter; a buffer for storing decoded data decoded by the data processing means; A compressed data decoding device having an analog signal converting means for reading decoded data stored in the storage device at a predetermined sampling rate and performing analog signal processing in accordance with a quality parameter; and a battery for supplying power; And a changing unit that changes the quality parameter so that at least one of the decoding process and the analog signal conversion process is in an operation state with low power consumption. . 6. A data processing means for decoding a data stream of a content which has been compression-encoded, and a de-quantization processing of the decoded data obtained thereby to obtain audio data, and a data processing means obtained by the data processing means. A compressed data decoding device comprising: a buffer for storing audio data; and an analog signal converting means for reading decoded data stored in the buffer at a predetermined sampling rate and performing analog signal processing in accordance with a quality parameter. A parameter determining means for determining a parameter according to a state or a data processing state for decoding the stream is provided, and the inverse quantization processing of the data processing means performs the bit of the encoded signal corresponding to the parameter determined by the parameter determining means. Compressed data decoding characterized by variable width No. device. 7. A data processing means for decoding a data stream of a content which has been compression-encoded, and for obtaining audio data by subjecting the decoded data thus obtained to inverse quantization, and data obtained by the data processing means. A compressed data decoding device comprising: a buffer for storing audio data; and analog signal conversion means for reading decoded data stored in the buffer at a predetermined sampling rate and performing analog signal processing in accordance with a quality parameter. Parameter determining means for determining a parameter in accordance with a state or a data processing state for decoding the stream, wherein the decoding processing in the data processing means is performed in a data stream decoding process corresponding to the parameter determined by the parameter determining means. Configuration that varies the data bit width A compressed data decoding device, characterized in that: 8. A data processing means for receiving a compression-encoded content data stream and performing a decoding process in accordance with a quality parameter, a buffer for storing decoded data decoded by the data processing means, and a buffer for storing the decoded data. An analog signal converting means for reading out the decoded data stored in the storage device at a predetermined sampling rate and performing analog signal processing according to the quality parameter, wherein the data processing for decoding the state or stream of the compressed data decoding device is performed. Determining means for determining a quality parameter in accordance with a situation, wherein the data processing means changes a sampling speed of the analog signal converting means in accordance with the quality parameter determined by the determining means; Decoding device. 9. The compressed data decoding apparatus according to claim 1, wherein said quality parameter determining means stores information on a speed difference between a decoding processing speed and a sampling speed in a state of free space in a data buffer. A compressed data decoding apparatus characterized in that a quality parameter is determined corresponding to the speed difference, obtained from a means for calculating the compressed data. 10. The compressed data decoding apparatus according to claim 1, wherein said quality parameter determining means obtains information on a speed difference between a decoding processing speed and a sampling speed, and decodes the decoding processing speed and the sampling speed for data. A compressed data decoding apparatus characterized in that a quality parameter is obtained from a calculating means for calculating from a speed difference between the compressed data and a speed parameter corresponding to the speed difference. 11. A data stream of a compressed and encoded content is received and decoded, the processed data is stored in a buffer, read out at a predetermined sampling rate, converted into an analog signal, and output. A compressed data decoding method applied to a compressed data decoding apparatus according to claim 1, wherein a quality parameter is determined, and the quality of data processing and analog signal conversion is changed according to the quality parameter. 12. A compression-encoded audio content data stream is received and decoded, the processed data is stored in a buffer, read out at a predetermined sampling rate, converted into an analog signal, and output. A decoding method applied to a compressed data decoding apparatus as described above, wherein a quality parameter is determined, and the quality of data processing and analog signal conversion is changed according to the quality parameter. 13. The compressed data decoding method according to claim 11, wherein the quality parameter is determined by comparing the decoding processing speed with the sampling speed of the analog signal conversion, and determining the quality parameter corresponding to the comparison result. And a decoding processing load is adjusted by changing the decoding and reproduction quality according to the determined quality parameter to prevent interruption of content reproduction.