JP2000138591A - Audio reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an audio reproducing device which can avoid the underflow and overflow of a bit buffer. SOLUTION: In an MPEG audio decoder 1, a control circuit 9 makes the generating periods of pipeline signals used for controlling the operations of sections 4-6 longer (or shorter) based on the discriminated result of an occupied amount discriminating circuit 4. When the periods are made longer (or shorter), the operating speeds of sections 6-8 become slower (or faster) and, since the reading-out speed of an audio stream from a bit buffer 2 also becomes slower (or faster), the underflow (or overflow) of the buffer 2 can be evaded. At the same time, a sound pitch converting circuit 5 makes the pitch of reproduced sounds nearly equal to that of normally reproduced sounds.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] The present invention relates to an audio reproducing apparatus.

[0002]

2. Description of the Related Art The information handled by multimedia is enormous and diverse, and it is necessary to process such information at high speed in order to put multimedia into practical use. In order to process information at high speed, data compression / decompression technology is indispensable. As such data compression / decompression technology, the “MPEG” method is exemplified. This MPEG system is based on the ISO (International Organization).
for Standardization) / IEC (Intarnational Elec)
trotechnical Commission)
It is being standardized by O / IEC JTC1 / SC29 / WG11).

[0003] MPEG is composed of three parts. Part 1 “MPEG System Part” (ISO / IEC
IS 11172 Part1: Systems) defines a multiplex structure (multiplex structure) of video data and audio data and a synchronization method. Part 2 "MPE
In the "G video part" (ISO / IEC IS 11172 Part2: Video), a high-efficiency encoding method of video data and a format of the video data are specified. Part 3, "MPEG
Audio Part ”(ISO / IEC IS 11172 Part3: Audio)
Defines a high-efficiency encoding method of audio data and a format of audio data.

At present, MPEG-1 and MPEG-1 are mainly used due to differences in encoding rates.
-2. MPEG-1 is mainly for video C
D (Compact Disc), CD-ROM (CD-Read Only Mem)
ory) and a storage medium using a recording medium such as a DVD (Digital Video Disk), and MPEG-2 supports a wide range of applications including MPEG-1.

A data string (bit stream) of video data encoded in accordance with the MPEG video part
Is called an MPEG video stream. Also, MP
A data string of audio data encoded according to the EG audio part is called an MPEG audio stream. Then, the video stream and the audio stream are time-division multiplexed in accordance with the MPEG system part and become an MPEG system stream as one data string.

[0006] MPEG audio has a layer (Laye).
r) There are three modes, I, Layer II and Layer III.
The higher the layer, the higher the sound quality and the higher the compression ratio. One frame of the audio stream is AAU (Audio Ac
cess Unit). The AAU is a minimum unit that can be decoded one by one, and includes data of a fixed number of samples (384 samples for Layer I, 1152 samples for Layer II and Layer III) set for each layer.

The AAU has a header and an optional error check (CRC; Cyclic Redundancy Co.) from the beginning.
de 16 bits), followed by audio data. The data from the header to the audio data is data used for reproducing the audio signal.

[0008] A sampling frequency is defined in the header. The sampling frequency is a field for designating a sampling rate, and includes three types of frequencies (32 kHz
, 44.1kHz, 48kHz).

The audio data is variable-length data. If the end of the audio data does not reach the end of the AAU, the remaining part is ancillary data (Ancillary data).
ary Data). This ancillary data includes
Any data other than MPEG audio can be inserted. In MPEG-2, multi-channel and multi-lingual data are inserted into ancillary data.

[0010] The audio data of layer I includes an allocation (Allocation) and a scale factor (Scale Fact).
or) and a sample. layer
II and layer III audio data are allocated, scale factor selection information (Scale Factor Slect
Information), scale factor, and sample.

The scale factor is a magnification at the time of reproducing the waveform for each sub-band and each channel, and is represented by 6 bits for each sub-band and channel, and is +6 to-.
Up to 118 dB can be specified in about 2 dB units. Since the value of the scale factor corresponds to the sound pressure level of the sound to be reproduced, when the value of the scale factor becomes lower than a certain level, the reproduced sound becomes a sound pressure level that cannot be heard by humans (ie, no sound).

The human auditory characteristics (acoustic psychological model) used in MPEG audio include a masking effect and a minimum audible characteristic. The masking effect means that when a loud sound is generated at a certain frequency, sounds below a certain level at frequencies near the loud sound become inaudible or hard to hear. The minimum audible characteristic is a certain frequency characteristic that the human ear is most sensitive to the human voice band of several hundred Hz, and sounds below a certain sound pressure level can not be heard in the ultra-low range or ultra-high range. It is to have.

Therefore, the masking effect and the minimum audible characteristic are combined to set a mask level that dynamically changes together with the audio signal, and a signal below this level is subjected to data compression. As a result, the encoding rate in layer I;
2k, 128kbps, compression ratio: 1/4, sound quality is CD-DA (CD Dig
ital Audio) and PCM (Pulse Code Modulation)
Equivalent to layer II, encoding rate; 128k, 96kbps
, Compression ratio: 1/6 to 1/8, sound quality is equivalent to MD and DCC, encoding rate for layer III: 128k, 96k, 64kbp
s, compression ratio; 1/6 to 1/12, a compression effect and sound quality can be obtained.

In the MPEG audio encoder, first, an input audio signal is divided into 32 sub-bands using a band division filter. Next, in the quantization, by using the masking effect and the minimum audibility characteristic as described above and by not assigning bits to the masked and inaudible speech, the amount of information is reduced and data compression is performed.

FIG. 3 shows a main block circuit of a conventional MPEG audio decoder 51.

The MPEG audio decoder 51 comprises a bit buffer 52 and a decode core circuit 53. The decode core circuit 53 includes an inverse quantization unit 54,
Each of the audio streams A, which comprises a band synthesis unit 55, a PCM output unit 56, and a control circuit 57, forms an
The AU (frame) is decoded according to the MPEG audio part.

The bit buffer 52 has a FIFO (First-In)
-First-Out) is composed of a ring buffer composed of a random access memory (RAM) and sequentially transfers audio streams transferred from external devices (recording media such as video CDs and DVDs, information devices such as personal computers, etc.). accumulate.

The control circuit 57 detects a header attached to the head of each AAU constituting the audio stream stored in the bit buffer 52 and, based on the detection result, outputs an audio stream for each AAU from the bit buffer 52. Is read. Further, the control circuit 57 detects a sampling frequency specified in the header, and generates a pipeline signal which is a pulse corresponding to the sampling frequency.

The operation of each section 54 to 56 is controlled according to a pipeline signal. That is, the operation speed of each section 54 to 56 corresponds to the pipeline signal.

The inverse quantization unit 54 performs inverse quantization of the quantization performed by the encoder for each AAU read from the bit buffer 52.

The band synthesizing unit 55 performs a product-sum operation by a butterfly operation on the output of the inverse quantization unit 54, and synthesizes the data divided into 32 sub-bands into one by the encoder.

The PCM output section 56 is composed of an output interface and a cross attenuator.
5, an audio signal (PCM output signal) is generated.

The audio signal is D / A converted by a D / A converter (not shown), and then amplified by an audio amplifier (not shown) and sent to a speaker (not shown). Then, the sound is reproduced from the speaker.

[0024]

As described above, the MPE
The data compressed in the G system is transferred from a recording medium such as a video CD or DVD, or an information device such as a personal computer to the MPEG audio decoder 51. Some have been transferred. In this case, when data cannot be received normally due to deterioration of the radio wave condition or when the reproduction frequency of the decode core circuit 53 is slightly higher than the bit rate of the audio stream transferred from the satellite, the data is reproduced in the bit buffer 52. There is no data to be provided, that is, an underflow occurs, and as a result, the sound of the reproduced sound may be interrupted and the user may find it difficult to hear.

On the other hand, when the reproduction frequency of the decode core circuit 53 is slightly slow, data to be reproduced is accumulated in the bit buffer 52 too much and cannot be accumulated, that is, overflow occurs. As a result, noise is generated in reproduced sound. The user may find it difficult to listen.

In information devices such as microcomputers, encoding of an audio stream is not always performed in accordance with the standard, and the bit rate of the audio stream may be out of the standard.
In this case, when the bit rate is slightly slower or faster than the reproduction frequency of the decode core circuit 53, the same problem as described above occurs.

The present invention has been made in order to solve the above problems, and an object of the present invention is to provide an audio reproducing apparatus which minimizes a user's difficulty in listening.

[0028]

According to a first aspect of the present invention, there is provided an audio reproducing apparatus comprising: a bit buffer for storing an audio stream; a decoding circuit for decoding each frame constituting the audio stream read from the bit buffer; And a decoding speed control circuit for changing the decoding speed of the decoding circuit based on the occupancy of the decoding circuit.

According to a second aspect of the present invention, there is provided an audio reproducing apparatus, comprising: a bit buffer for storing an audio stream; a decoding circuit for decoding each frame constituting the audio stream read from the bit buffer in accordance with an MPEG audio part; An occupancy determination circuit that detects the occupancy of the buffer and compares the occupancy with a predetermined threshold, and a decode speed control circuit that changes the decoding speed of the decoding circuit based on the determination result of the occupancy determination circuit The gist is to have

That is, when the amount of data stored in the bit buffer and the amount of data decoded by the decoding circuit are unbalanced, the decoding speed of the decoding circuit is changed to balance both data amounts. .

When the occupancy of the bit buffer is smaller than a predetermined first threshold value, the occupancy determination circuit sends a signal notifying that the bit buffer is in an underflow state to the decoding speed control circuit. Send out,
It is desirable that the decoding speed control circuit be configured to reduce the decoding speed of the decoding circuit based on this signal.

Further, an audio pitch conversion circuit for increasing the pitch of reproduced audio may be provided in response to a decrease in the decoding speed of the decoding circuit. In other words, when the decoding speed of the decoding circuit decreases, the pitch (pitch) of the reproduced sound decreases, and the utterance speed (speech speed) decreases, so that the voice pitch conversion circuit corrects this state. Then, the pitch of the sound to be reproduced is made substantially the same as in normal reproduction.

When the occupancy of the bit buffer is larger than a second predetermined threshold, the occupancy determination circuit outputs a signal indicating that the bit buffer is in an overflow state to the decode speed control circuit. It is preferable that the decoding speed control circuit increases the decoding speed of the decoding circuit based on this signal.

In this case, an audio pitch conversion circuit for lowering the pitch of reproduced audio may be provided in response to an increase in the decoding speed of the decoding circuit. That is, when the decoding speed of the decoding circuit increases, the pitch (pitch) of the reproduced sound increases, and in addition, the utterance speed (speech speed) increases.
Becomes faster, the voice pitch conversion circuit makes the pitch of the reproduced voice almost the same as in normal reproduction in order to correct this state.

[0035]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) Hereinafter, a first embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows a main part block circuit of the MPEG audio decoder 1 of the present embodiment.

The MPEG audio decoder 1 comprises a bit buffer 2, a decode core circuit 3, an occupancy determination circuit 4,
It comprises a voice pitch conversion circuit 5. The circuits 2 to 5 are mounted on a one-chip LSI.

The decoding core circuit 3 is composed of an inverse quantization unit 6, a band synthesis unit 7, a PCM output unit 8, and a control circuit 9, and each of the audio signals A constituting the MPEG audio stream.
The AU (frame) is decoded according to the MPEG audio part.

As will be described later, an audio stream transferred from a digital TV satellite is sent to a bit buffer 2.

The bit buffer 2 is a FIFO RAM.
, And sequentially stores audio streams.

The control circuit 9 detects a header attached to the head of each AAU constituting the audio stream stored in the bit buffer 2 and, based on the detection result, outputs an audio stream corresponding to one AAU from the bit buffer 2. Is read. Further, the control circuit 9 detects a sampling frequency specified in the header, and generates a pipeline signal which is a pulse corresponding to the sampling frequency.

The operation of each section 6 to 8 is controlled according to a pipeline signal. That is, the operation speed of each section 6 to 8 corresponds to the pipeline signal.

The inverse quantization unit 6 performs inverse quantization on each AAU read from the bit buffer 2.

The band synthesizing unit 7 performs a product-sum operation by a butterfly operation on the output of the inverse quantization unit 6, and synthesizes the data divided into 32 sub-bands into one.

The PCM output section 8 includes an output interface and a cross attenuator, and generates an audio signal (PCM output signal) from the output of the band synthesis section 7.

The occupancy determination circuit 4 detects the occupancy Q of the bit buffer 2, determines whether there is a possibility of underflow, and transmits the result to the control circuit 9.

The voice pitch conversion circuit 5 increases or decreases the bit rate of the decoded audio signal. Based on such a configuration, the respective operations when the underflow occurs in the bit buffer 2 and when the overflow occurs will be described below. (When underflow occurs) For digital TV broadcasting, an MPEG audio stream is input from a satellite to a bit buffer 2 and sequentially transferred to a decoding core circuit 3 to perform decoding processing.

At this time, when data cannot be received normally due to deterioration of the radio wave condition or when the reproduction frequency (generation cycle of the pipeline signal) of the decode core circuit 3 is changed with respect to the bit rate of the audio stream transferred from the satellite.
Is slightly faster, the amount of data decoded by the decode core circuit 3 is larger than the amount of data stored in the bit buffer 2, and there is no data to be reproduced in the bit buffer 2, that is, A flow will occur. When an underflow occurs in the bit buffer 2, the reproduced sound may be interrupted, and the user may find it difficult to hear.

Therefore, the occupancy determining circuit 4 determines that the bit buffer 2 may underflow when the occupancy Q becomes smaller than a predetermined threshold TH1. The control circuit 9 increases the generation cycle of the pipeline signal based on the determination result of the occupancy determination circuit 4 while the occupancy Q is lower than the threshold value TH1. Since the operation of each unit 6 to 8 is controlled according to the pipeline signal, each unit 6 to 8
The operation speed 8 corresponds to the generation cycle of the pipeline signal. Therefore, when the generation cycle of the pipeline signal becomes longer, the operation speed of each unit 6 to 8 becomes slower.

The speed at which the audio stream is read from the bit buffer 2 depends on the processing speed of the decode core circuit 3 (ie, the operating speed of each section 6 to 8). Therefore, if the operation speed of each of the units 6 to 8 is reduced, the speed at which the audio stream is read from the bit buffer 2 is also reduced, so that the underflow of the bit buffer 2 can be avoided.

Therefore, it is possible to prevent the reproduced sound from being hard to hear due to the underflow of the bit buffer 2.

However, the bit rate of the audio signal decreases as the processing speed of the decode core circuit 3 decreases. As a result, the pitch (pitch) of the reproduced sound is lowered, and the utterance speed (speech speed) is lowered. Therefore, although the sound is not interrupted in the reproduced sound, it may be hard to hear in some cases.

The voice pitch conversion circuit 5 converts the bit rate of the audio signal to the same value as the normal value in conjunction with the control circuit 9 increasing the generation cycle of the pipeline signal in order to correct this state. By doing so, the pitch of the reproduced sound is made almost the same as in normal reproduction.

Then, when the occupation amount Q becomes larger than the threshold value TH1 again, the occupation amount determination circuit 4 determines that there is no possibility that the bit buffer 2 will underflow.
In this case, the control circuit 9 detects the sampling frequency specified in the header of each AAU again, and outputs a pipeline signal corresponding to the sampling frequency to each of the units 6 to 8.
Control the operation of. At the same time, the conversion operation by the voice pitch conversion circuit 5 ends. On the other hand, when data cannot be received normally due to deterioration of the radio wave condition, or when the bit rate of the audio stream transferred from the satellite is higher than the reproduction frequency of the decode core circuit 3 (the generation cycle of the pipeline signal). ) Is slightly slower, the amount of data stored in the bit buffer 2 is larger than the amount of data decoded by the decode core circuit 3, and the data to be reproduced in the bit buffer 2 is too large to be stored. A condition, ie, overflow, will occur. When an overflow occurs in the bit buffer 2, noise may be generated in the reproduced sound, and the user may feel uncomfortable.

Therefore, the occupancy determining circuit 4 determines that the bit buffer 2 may overflow when the occupancy Q becomes larger than a predetermined threshold TH2. The control circuit 9 shortens the generation cycle of the pipeline signal based on the determination result of the occupancy determination circuit 4 while the occupancy Q exceeds the threshold TH2. Since the operation speed of each of the units 6 to 8 corresponds to the generation cycle of the pipeline signal, the operation speed of each of the units 6 to 8 increases as the generation period of the pipeline signal becomes shorter.

Therefore, if the operation speed of each of the units 6 to 8 is increased, the speed at which the audio stream is read out from the bit buffer 2 is also increased, so that the overflow of the bit buffer 2 can be avoided.

Therefore, it is possible to prevent the reproduction sound from being hard to hear due to the overflow of the bit buffer 2.

However, as the processing speed of the decode core circuit 3 increases, the bit rate of the audio signal increases. As a result, in addition to increasing the pitch (pitch) of the reproduced sound, the utterance speed (speaking speed) increases. Therefore, although generation of noise in the reproduced sound is suppressed, in some cases, it may be hard to hear.

In order to correct this state, the voice pitch conversion circuit 5 converts the bit rate of the audio signal to the same value as usual in conjunction with the control circuit 9 shortening the generation cycle of the pipeline signal. By doing so, the pitch of the reproduced sound is made almost the same as in normal reproduction.

When the occupancy Q becomes smaller than the threshold value TH2 again, the occupancy determination circuit 4 determines that there is no possibility that the bit buffer 2 will overflow.
In this case, the control circuit 9 detects the sampling frequency specified in the header of each AAU again, and outputs a pipeline signal corresponding to the sampling frequency to each of the units 6 to 8.
Control the operation of. At the same time, the conversion operation by the voice pitch conversion circuit 5 ends.

Incidentally, data sent from a digital TV satellite also includes an MPEG video stream. As described above, the video stream and the audio stream are time-division multiplexed in accordance with the MPEG system part. And sent as an MPEG system stream as one data string.

For such data, an MPEG video decoder is also required in addition to the MPEG audio decoder 1. FIG.
3 shows a block circuit of an EG system decoder.

The MPEG system decoder 21 includes an audio video parser (AV parser) 22, an MPEG video decoder 23, and the MPEG audio decoder 1.

The AV parser 22 includes a demultiplexer (D
MUX; DeMUltipleXer) 24 for inputting an MPEG system stream transferred from a digital TV satellite. The DMUX 24 separates the system stream into an MPEG video stream and an MPEG audio stream. The video stream is output to the video decoder 23, and the audio stream is output to the audio decoder 1.

The video decoder 23 decodes a video stream in accordance with the MPEG video part and generates a video signal. The video signal is output to the display 25, and the display 25 reproduces a moving image.

The audio decoder 1 generates an audio signal as described above, and the audio signal is D / A
After being D / A-converted by the converter 26, it is amplified by the audio amplifier 27 and sent to the speaker 28. Then, the sound is reproduced from the speaker 28.

In the above embodiment, each circuit (2 to 2)
The signal processing in 9) may be replaced with software-based signal processing using a CPU.

[0068]

As described above in detail, according to the present invention, there is provided an audio reproducing apparatus capable of avoiding underflow or overflow of a bit buffer and preventing a user from feeling uncomfortable as much as possible. Can be.

[Brief description of the drawings]

FIG. 1 is a main part block circuit diagram of an embodiment embodying the present invention.

FIG. 2 is a main part block circuit diagram of an MPEG system decoder.

FIG. 3 is a main part block circuit diagram of a conventional mode.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 MPEG audio decoder 2 Bit buffer 3 Decoding core circuit 4 Occupancy amount determination circuit 5 Audio pitch conversion circuit 6 Inverse quantization 7 Band synthesis unit 8 PCM output unit 9 Control circuit

Claims (6)

[Claims] 1. A bit buffer for storing an audio stream, a decoding circuit for decoding each frame constituting the audio stream read from the bit buffer, and a decoding speed of the decoding circuit based on an occupancy of the bit buffer. An audio reproducing apparatus comprising: a decoding speed control circuit for changing the decoding speed. 2. A bit buffer for storing an audio stream, a decoding circuit for decoding each frame constituting the audio stream read from the bit buffer in accordance with an MPEG audio part, and detecting an occupancy of the bit buffer. An occupancy determination circuit that compares the occupancy with a predetermined threshold, and a decoding speed control circuit that changes the decoding speed of the decoding circuit based on the determination result of the occupancy determination circuit. Characteristic audio playback device. 3. The decoding rate control circuit according to claim 1, wherein said occupancy determination circuit outputs a signal indicating that the bit buffer is in an underflow state when the occupancy of the bit buffer is smaller than a predetermined first threshold. 3. The audio reproducing apparatus according to claim 1, wherein the signal is sent to a circuit, and based on the signal, the decoding speed control circuit reduces the decoding speed of the decoding circuit. 4. The audio reproduction according to claim 1, further comprising an audio pitch conversion circuit for increasing a pitch of reproduced audio in response to a decrease in the decoding speed of the decoding circuit. apparatus. 5. The decoding speed control circuit according to claim 1, wherein the occupancy determination circuit outputs a signal indicating that the bit buffer is in an overflow state when the occupancy of the bit buffer is larger than a second predetermined threshold value. 3. The audio reproducing apparatus according to claim 1, wherein the decoding speed control circuit increases the decoding speed of the decoding circuit based on the signal. 6. The audio reproduction according to claim 1, further comprising an audio pitch conversion circuit for lowering a pitch of reproduced audio in response to an increase in the decoding speed of the decoding circuit. apparatus.