JPH10178349A - Coding and decoding method for audio signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To encode lots of additional information simultaneously in the case of coding a digital audio signal. SOLUTION: Relating to the coding method for an audio signal generating an audio bit stream according to an MPEG 1 audio layer 1 format, a header 11, an error check 12 and audio data 13 consisting of only bit assignment, and at least one or more frames consisting of ancillary data 14 are included, then lots of additional information is coded in the ancillary data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio signal encoding method and an audio signal decoding method.

[0002]

2. Description of the Related Art In recent years, digital audio signal encoding methods have been actively developed. Particularly, as a high quality audio encoding method, MPEG1 audio encoding (ISO / IEC) has been proposed.
11172-3) is known.

FIG. 9 is a block diagram showing the basic structure of an encoder for MPEG1 audio encoding. When audio samples are input to the encoder 91, the mapping unit 92 filters the input audio samples and generates a thinned representation. The psychoacoustic model unit 93 generates a data set that controls quantization and encoding. The quantization and encoding unit 94 includes a layer 1, a layer 2 and a layer 3
Is performed according to any of the encoding methods. The frame assembling unit 95 assembles an actual bit stream from the output data of the quantization and encoding unit 94 and the ancillary data, and adds other information (eg, error check) as necessary.

FIG. 10 is a block diagram showing the basic structure of a decoder for MPEG1 audio encoding. When the bit stream is input to the decoder 101, the frame
02 separates the bitstream to recover various parts of the information. The restoration unit 103 restores a quantized form of a series of mapping samples. The inverse mapping unit 104 returns these mapping samples to PCM audio samples.

[0005] An audio bit stream generated and output by an encoder or input to a decoder and reproduced is composed of continuous frames, and is composed of ISO / IEC 11172-3.
When described in the format described in

A frame is composed of a header, an error check, audio data and ancillary data.
Similarly, the description of a frame in the format described in ISO / IEC 11172-3 is as follows.

[0007] The header contains information on the synchronization word and the encoding method. The error check is a bit for error check. The audio data is audio data encoded by any of Layer 1, Layer 2, and Layer 3 methods. Ancillary data consists of ancillary bits describing additional information other than audio data. However, in the case of Layer 3, ancillary bits are included in audio data.

When a header, an error check, and an ancillary data are described in a frame in a format described in ISO / IEC 11172-3, the following is obtained. Header: header () ｛syncword 12 bits ID 1 bit layer 2 bits protection_bit 1 bit bitrate_index 4 bits sampling_frequency 2 bits padding _bit 1 bit private _bit 1 bit mode 2 bits mode_extension 2 bits copyright 1 bit original / copy 1 bit emphasis 2 bits ｝ Error check: error_check () ｛if (protection _bit ==) crc_check 16 bits｝ Ancillary data: ancillary _data () ｛if ((layer == 1) || (layer == 2)) for (b = 0; b <no_of_ancillary_bits; b ++) ancillary_bit 1 bit｝

The bit streams of Layer 1, Layer 2 and Layer 3 are schematically shown in FIG. 11, FIG. 12 and FIG.

FIG. 11 shows a bit stream format of the MPEG1 audio encoding layer 1. A bit stream is composed of one or more consecutive frames. FIG. 11 shows a format of one frame. The frame is composed of a header 111, an error check 112, bit allocation as audio data 113, scale factors and samples, and ancillary data 114.

FIG. 12 shows a bit stream format of the MPEG1 audio encoding layer 2. Similarly,
A bit stream is composed of one or more consecutive frames. FIG. 12 shows a format of one frame.
The frame includes a header 121, an error check 122, bit allocation as audio data 123, scale factor selection information, scale factors and samples, and ancillary data 124.

FIG. 13 shows a bit stream format of the MPEG1 audio encoding layer 3. Similarly,
A bit stream is composed of one or more consecutive frames. FIG. 13 shows a format of one frame.
The frame includes a header 131, an error check 132, and additional information such as private bits described as audio data 133, scale factor selection information, quantization step width information, and main information such as scale factors, Huffman encoded data, and ancillary bits. Consists of

[0013]

However, according to the conventional method, if information other than audio data such as character information is to be included in a bit stream, the information in the ancillary data (audio data in the case of layer 3) is not included. Since encoding is performed as ancillary bits, audio data and information other than the audio data coexist in the same frame in the bit stream. Therefore,
There is a problem that the bit rate allowed for the audio data is reduced, and the sound quality is deteriorated as compared with the case where all the allowable bit rates are used for the audio data.

SUMMARY OF THE INVENTION The present invention is to solve such a conventional problem. An audio signal capable of encoding a sufficient amount of information other than audio data while maintaining the quality of audio data. It is an object of the present invention to provide an encoding method and a decoding method.

[0015]

In order to achieve the above object, the present invention provides an encoding method in which a bit stream is composed of a frame composed of audio data and a frame composed of information other than audio data. Things. According to the present invention, it is possible to encode not only audio data of high quality but also a sufficient amount of information other than audio data, for example, character information relating to audio data to be encoded.

Further, in order to achieve the above object, the present invention provides a decoding method in which a bit stream is constituted by a frame composed of audio data and a frame composed of information other than audio data. According to the present invention, it is possible to decode not only audio data of high quality but also a sufficient amount of information other than audio data, for example, character information relating to audio data to be decoded.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An audio signal encoding method according to claim 1 of the present invention is an audio signal encoding method for generating an audio bit stream according to the MPEG1 audio layer 1 format. At least one frame including only a header, an error check, bit allocation, and ancillary data is included.
The PEG1 audio layer 1 encoding method has an effect that information other than audio data of a sufficient capacity can be encoded without deteriorating the quality of audio data.

According to a second aspect of the present invention, there is provided a method for encoding an audio signal for generating an audio bit stream according to the MPEG1 audio layer 2 format, comprising the steps of:
At least one frame composed of only bit allocation and ancillary data is included. In the encoding method of MPEG1 audio layer 2, without deteriorating the quality of audio data,
This has an effect that information other than audio data having a sufficient capacity can be encoded.

According to a third aspect of the present invention, there is provided a method for encoding an audio signal for generating an audio bit stream according to the MPEG1 audio layer 3 format, comprising the steps of:
At least one frame composed of only private bits and ancillary bits is included. In the MPEG1 audio layer 3 encoding method, audio data of sufficient capacity can be obtained without deteriorating the quality of audio data. It has an effect that information other than the above can be encoded.

According to a fourth aspect of the present invention, there is provided a method for encoding an audio signal for generating an audio bit stream according to the MPEG1 audio layer 1 format, wherein the bit stream comprises only a header, bit allocation and ancillary data. At least one frame.
The encoding method of the G1 audio layer 1 has an effect that information other than audio data having a sufficient capacity can be encoded without deteriorating the quality of audio data.

According to a fifth aspect of the present invention, there is provided a method for encoding an audio signal for generating an audio bit stream according to the MPEG1 audio layer 2 format, wherein the bit stream comprises only a header, bit allocation and ancillary data. At least one frame.
The encoding method of the G1 audio layer 2 has an effect that information other than audio data with a sufficient capacity can be encoded without deteriorating the quality of audio data.

According to a sixth aspect of the present invention, there is provided a method of encoding an audio signal for generating an audio bit stream according to the MPEG1 audio layer 3 format, wherein the bit stream comprises only a header, a private bit, and an ancillary bit. At least one frame is included,
In the encoding method of MPEG1 audio layer 3,
This has the effect that information other than audio data having a sufficient capacity can be encoded without deteriorating the quality of the audio data.

According to a seventh aspect of the present invention, there is provided a method for decoding an audio signal for reproducing an audio bit stream in accordance with the MPEG1 audio layer 1 format, comprising the steps of: The MPEG1 audio layer 1 decoding method decodes information other than audio data having a sufficient capacity without deteriorating the quality of audio data. Has the effect of being able to.

An eighth aspect of the present invention is a decoding method of an audio signal for reproducing an audio bit stream conforming to the MPEG1 audio layer 2 format, wherein a header, an error check, a bit allocation and an ancillary data are included in the bit stream. The MPEG1 audio layer 1 decoding method decodes information other than audio data having a sufficient capacity without deteriorating the quality of audio data. Has the effect of being able to.

According to a ninth aspect of the present invention, in the decoding method of an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 3 format, the bit stream includes only a header, an error check, a private bit, and an ancillary bit. The MPEG1 audio layer 3 decoding method can decode information other than audio data having a sufficient capacity without deteriorating the quality of audio data. It has the action of:

According to a tenth aspect of the present invention, in the method for decoding an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 1 format, the bit stream includes only a header, bit allocation, and ancillary data. The MPE contains at least one frame.
The method for decoding the G1 audio layer 1 has an effect that information other than audio data having a sufficient capacity can be decoded without deteriorating the quality of audio data.

According to an eleventh aspect of the present invention, in the method for decoding an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 2 format, the bit stream includes only a header, bit allocation, and ancillary data. The MPE contains at least one frame.
The method of decoding the G1 audio layer 2 has an effect that information other than audio data of a sufficient capacity can be decoded without deteriorating the quality of audio data.

According to a twelfth aspect of the present invention, in the decoding method of an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 3 format, a frame comprising only a header, a private bit and an ancillary bit in the bit stream. At least one is included,
The method of decoding the MPEG1 audio layer 3 has an effect that information other than audio data having a sufficient capacity can be decoded without deteriorating the quality of audio data.

Hereinafter, an embodiment of the present invention will be described with reference to FIG.
This will be described with reference to FIG. First, the encoding method will be described, and then the decoding method will be described. (Embodiment 1) FIG. 1 shows a bit stream format of an MPEG1 audio encoding layer 1. A bit stream is composed of one or more consecutive frames. FIG. 1 describes only the format of one frame. The frame includes a header 11, an error check 12, audio data 13, and ancillary data 14. Audio data 13 consists only of bit assignments. This is possible in the following cases.

The case of the layer 1 stereo channel will be described. FIG. 2 shows the contents of the bit allocation information. The bit allocation information of the stereo channel needs to be specified for the 32 sub-bands of each of the L and R channels. As for bit allocation information, 4 bits are allocated to each band. By setting this value to 0,
The number of bits for quantizing each subband becomes 0, and the samples are not quantized. In this case, the scale factor information is not quantized. The number of bits used for bit allocation is 4 bits × 32 subbands × 2 channels =
256 bits.

For example, the sampling frequency of an input audio sample is 48 kHz, and the bit rate is 256 kb.
Suppose it is encoded with it / s. One frame of layer 1 is 38
Since information of four samples is included, the number of bits allowed in one frame of the bit stream is 2048 bits.
Since the number of bits used for the header is 32 bits and the number of bits used for error checking is 16 bits, the number of bits usable for ancillary data is 2048− (32+
16 + 256) = 1754 bits. Assuming that kanji is written in ancillary data at 16 bits / character, 1
09 characters can be written.

The operation of the encoder will be described with reference to FIG. A level 0, that is, a silent audio signal is input to the mapping unit 92 of the encoder 91. The psychoacoustic model unit 93 generates a data set for controlling quantization and encoding. Since the input audio samples are silent, the bit allocation information for each subband is determined to be zero. Therefore, the quantization and encoding unit 94 does not quantize and encode the audio samples. As a result, only the header, error check, and bit allocation are sent to frame assembler 95. On the other hand, ancillary data such as kanji is sent to the frame assembling unit 95. The frame is generated by the frame assembling unit 95 and output as an encoded bit stream. Subsequently, when a sounded audio signal is input to the mapping unit 92, it is encoded as a frame including audio data. Also,
The frame described in the bit stream format of FIG. 1 may be located at any position in the bit stream. That is, the audio signal may be mixed with the encoded frame.

(Embodiment 2) FIG. 3 shows a bit stream format of the MPEG1 audio encoding layer 2. A bit stream is composed of one or more consecutive frames. FIG. 3 describes only the format of one frame. The frame includes a header 31, an error check 32, audio data 33, and ancillary data 34. Audio data 33 consists only of bit assignments. This is possible in the following cases.

The case of the layer 2 stereo channel will be described. FIG. 4 shows the contents of the bit allocation information. It is necessary to specify the bit allocation information of the stereo channel for the band-divided subbands of the L and R channels. The number of bits used for the bit allocation information is determined in advance for each sampling frequency and bit rate of an input audio sample. FIG. 4 shows a case where the sampling frequency is 48 kHz and the bit rate is 256 kbit / s. By setting the bit allocation information of each subband to 0, the number of bits for quantizing each subband becomes 0, and the sample is not quantized. In this case, the scale factor information and the scale factor selection information are not quantized. The number of bits used for bit allocation is a sampling frequency of 48 kHz and a bit rate of 256 kbit / s.
In the case of (4 bits × 11 subbands + 3 bits × 12
Subband + 2 bits × 4 subbands) × 2 channels =
176 bits.

In the layer 2, one frame contains information of 1152 samples. Therefore, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the number of bits allowed for one frame of the bit stream is 6144 bits. Since the number of bits used for the header is 32 bits and the number of bits used for error checking is 16 bits, the number of bits usable for ancillary data is 6144- (32 + 1).
6 + 176) = 5920 bits. If you write kanji in ancillary data at 16 bits / character,
370 characters can be written.

The operation of the encoder is the same as that of layer 1 of the first embodiment.
Is the same as

(Embodiment 3) FIG. 5 shows a bit stream format of the MPEG1 audio encoding layer 3. A bit stream is composed of one or more consecutive frames. FIG. 5 describes only the format of one frame. The frame includes a header 51, an error check 52, and audio data 53. The audio data 53 includes only a main information start code, a private bit, and an ancillary bit.

The case of the layer 3 stereo channel will be described. In layer 3, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the bit stream 1
The number of bits allowed for a frame is 6
It is 144 bits. The number of bits used for the header is 3
Since 2 bits and 16 bits are used for error checking, the number of bits usable for audio data is 6144- (32 + 16) = 6096 bits. Since 9 bits are assigned to the main information start code and 3 bits (5 bits in the case of the single channel mode) to the private bit in the audio data, 6096− (9 + 3) = 6084 bits are assigned to the ancillary bits. 16-bit kanji for ancillary bit /
When writing with characters, 380 characters can be written.

The operation of the encoder will be described with reference to FIG. A level 0, that is, a silent audio signal is input to the mapping unit 92 of the encoder 91. The psychoacoustic model unit 93 generates a data set for controlling quantization and encoding. Since the input audio sample is in a silent state, the quantization and encoding section 94 does not quantize and encode the audio sample. As a result, only the header, error check, main information start code, and private bit are sent to the frame assembling unit 95. On the other hand, kanji and other ancillary bits are
Sent to The frame is generated by the frame assembling unit 95 and output as an encoded bit stream. Subsequently, when a sounded audio signal is input to the mapping unit 92, it is encoded as a frame including audio data. . Further, the frame described in the bit stream format of FIG. 5 may be located at an arbitrary position in the bit stream. That is, the audio signal may be mixed with the encoded frame.

(Embodiment 4) FIG. 6 shows a bit stream format of the MPEG1 audio encoding layer 1. A bit stream is composed of one or more consecutive frames. FIG. 6 describes only the format of one frame. The frame includes a header 61, audio data 62, and ancillary data 63. Audio data 62 consists only of bit allocation information. In the case of a layer 1 stereo channel, as in (Embodiment 1), the number of bits used for bit allocation is 4 bits ×
32 subbands × 2 channels = 256 bits.

For example, if encoding is performed at a bit rate of 256 kbit / s, if the sampling frequency of an input audio sample is 48 kHz, one bit of the bit stream is output.
The number of bits allowed for a frame is 2048 bits. Since the number of bits used for the header is 32 bits,
The number of bits that can be used for ancillary data is 2048-
(32 + 256) = 1760 bits. If kanji is written in the ancillary data at 16 bits / character, 110 characters can be written.

The operation of the encoder will be described with reference to FIG. A level 0, that is, a silent audio signal is input to the mapping unit 92 of the encoder 91. The psychoacoustic model unit 93 generates a data set for controlling quantization and encoding. Since the input audio samples are silent, the bit allocation information for each subband is determined to be zero. Therefore, the quantization and encoding unit 94 does not quantize and encode the audio samples. As a result, only the header and bit allocation are sent to the frame assembling unit 95. On the other hand, ancillary data such as kanji is sent to the frame assembling unit 95. Frame assembly part 95
Generates a frame and outputs the frame as an encoded bit stream. Subsequently, when the audio signal in a sound state is input to the mapping unit 92, it is encoded as a frame including the audio signal. Alternatively, the frame described in the bit stream format of FIG. 6 may be at an arbitrary position in the bit stream. That is,
The audio signal may be mixed with the encoded frame.

(Embodiment 5) FIG. 7 shows a bit stream format of the MPEG1 audio encoding layer 2. A bit stream is composed of one or more consecutive frames. FIG. 7 describes only the format of one frame. The frame includes a header 71, audio data 72, and ancillary data 73. Audio data 72 consists only of bit allocation. Layer 2
, The number of bits used for bit allocation is 176 bits, as in the second embodiment.

For example, when encoding is performed at a bit rate of 256 kbit / s, if the sampling frequency of an input audio sample is 48 kHz, the number of bits allowed for one frame of a bit stream is 6144 bits. Since the number of bits used for the header is 32 bits, the number of bits usable for ancillary data is 2048−32 = 2
This is 016 bits. If kanji is written in the ancillary data at 16 bits / character, 126 characters can be written.

The operation of the encoder is the same as that of the layer 1 of the fourth embodiment.
Is the same as

(Embodiment 6) FIG. 8 shows a bit stream format of the MPEG1 audio encoding layer 3. A bit stream is composed of one or more consecutive frames. FIG. 8 describes only the format of one frame. The frame includes a header 81 and audio data 82. The audio data 82 includes only a main information start code, a private bit, and an ancillary bit.

The case of the layer 3 stereo channel will be described. In layer 3, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the bit stream 1
The number of bits allowed for a frame is 6144 bits. Since the number of bits used for the header is 32 bits,
The number of bits available for audio data is 6144-
32 = 6112 bits. 9 bits for main information start code and 3 for private bits in audio data
Since bits (5 bits in single channel mode) are allocated, 6112-
(9 + 3) = 6100 bits are allocated. If kanji is written at 16 bits / character in the ancillary bit, 381 characters can be written.

The operation of the encoder will be described with reference to FIG. A level 0, that is, a silent audio signal is input to the mapping unit 92 of the encoder 91. The psychoacoustic model unit 93 generates a data set for controlling quantization and encoding. Since the input audio sample is in a silent state, the quantization and encoding section 94 does not quantize and encode the audio sample. As a result, only the header, the main information start code, and the private bit are sent to the frame assembling unit 95. On the other hand, ancillary bits such as kanji are also sent to the frame assembling section 95. The frame is generated by the frame assembling unit 95 and output as an encoded bit stream. Subsequently, when a sounded audio signal is input to the mapping unit 92, it is encoded as a frame including audio. Further, the frame described in the bit stream format of FIG.
It may be at any position in the bitstream. That is, the audio signal may be mixed with the encoded frame.

(Embodiment 7) Next, a decoding method of the present invention will be described. Naturally, the following decoding methods correspond to the above-described encoding methods.
That is, Embodiment 7 corresponds to Embodiment 1, Embodiment 8 corresponds to 2, Embodiment 9 corresponds to 3, and Embodiment 10 corresponds to
Corresponds to 4, the eleventh embodiment corresponds to 5, and the twelfth embodiment corresponds to 6.

FIG. 1 shows a bit stream format of the MPEG1 audio encoding layer 1. A bit stream is composed of one or more consecutive frames.
FIG. 1 describes only the format of one frame. The frame includes a header 11, an error check 12, audio data 13, and ancillary data 14. Audio data 13 consists only of bit assignments. This is possible in the following cases.

The case of the layer 1 stereo channel will be described. FIG. 2 shows the contents of the bit allocation information. The bit allocation information of the stereo channel needs to be specified for the 32 sub-bands of each of the L and R channels. As for bit allocation information, 4 bits are allocated to each band. By setting this value to 0,
The number of bits for quantizing each subband becomes 0, and the samples are not quantized. In this case, the scale factor information is not quantized. The number of bits used for bit allocation is 4 bits × 32 subbands × 2 channels =
256 bits.

For example, the sampling frequency of the input audio sample is 48 kHz, and the bit rate is 256 kb.
Suppose it is encoded with it / s. Since one frame of layer 1 includes information of 384 samples, the number of bits allowed for one frame of the bit stream is 2048 bits. Since the number of bits used for the header is 32 bits and the number of bits used for error checking is 16 bits, the number of bits usable for ancillary data is 2048−
(32 + 16 + 256) = 1754 bits. Assuming that Chinese characters are written in the ancillary data at 16 bits / character, 109 characters have been written.

The operation of the decoder will be described with reference to FIG.
When the bit stream is input to the frame decomposing unit 102 of the decoder 101, the frame decomposing unit 102 performs an error check, separates audio data and ancillary data, and outputs the audio data to the decompression unit 103. Is output to the outside. The audio data is sent to the restoration unit 103, where the audio data is dequantized, and
At 04, it is inverse filtered and output as PCM audio samples. The ancillary data output to the outside is converted into appropriate information by an external device. For example, if it is character information, it can be displayed on a display. If there is a frame in which the audio signal is encoded following the frame described in the bit stream format of FIG. 1, the audio signal can be decoded and output almost at the same time as the characters are displayed on the display. become. The frame described in the bit stream format of FIG.
It can be at any position in the bitstream.

(Embodiment 8) FIG. 3 shows a bit stream format of the MPEG1 audio encoding layer 2. A bit stream is composed of one or more consecutive frames. FIG. 3 describes only the format of one frame. The frame includes a header 31, an error check 32, audio data 33, and ancillary data 34. Audio data 33 consists only of bit assignments. This is possible in the following cases.

The case of the layer 2 stereo channel will be described. FIG. 4 shows the contents of the bit allocation information. It is necessary to specify the bit allocation information of the stereo channel for the band-divided subbands of the L and R channels. The number of bits used for the bit allocation information is determined in advance for each sampling frequency and bit rate of an input audio sample. FIG. 4 shows a case where the sampling frequency is 48 kHz and the bit rate is 256 kbit / s. By setting the bit allocation information of each subband to 0, the number of bits for quantizing each subband becomes 0, and the sample is not quantized. In this case, the scale factor information and the scale factor selection information are not quantized. The number of bits used for bit allocation is a sampling frequency of 48 kHz and a bit rate of 256 kbit / s.
In the case of (4 bits × 11 subbands + 3 bits × 12
Subband + 2 bits × 4 subbands) × 2 channels =
176 bits.

In the layer 2, one frame contains information of 1152 samples. Therefore, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the number of bits allowed for one frame of the bit stream is 6144 bits. Since the number of bits used for the header is 32 bits and the number of bits used for error checking is 16 bits, the number of bits usable for ancillary data is 6144- (32 + 1).
6 + 176) = 5920 bits. If kanji is written in the ancillary data at 16 bits / character, 370 characters have been written.

The operation of the decoder is the same as that of the layer 1 of the seventh embodiment.
Is the same as

(Embodiment 9) FIG. 5 shows the bit stream format of the MPEG1 audio encoding layer 3. A bit stream is composed of one or more consecutive frames. FIG. 5 describes only the format of one frame. The frame includes a header 51, an error check 52, and audio data 53. The audio data 53 includes only a main information start code, a private bit, and an ancillary bit.

The case of the layer 3 stereo channel will be described. In layer 3, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the bit stream 1
The number of bits allowed for a frame is 6
It is 144 bits. The number of bits used for the header is 3
Since 2 bits and 16 bits are used for error checking, the number of bits usable for audio data is 6144- (32 + 16) = 6096 bits. Since 9 bits are assigned to the main information start code and 3 bits (5 bits in the case of the single channel mode) to the private bit in the audio data, 6096− (9 + 3) = 6084 bits are assigned to the ancillary bits. If you write kanji in the ancillary bit, you need 16 bits for each kanji character.
This means that 80 characters have been written.

The operation of the decoder will be described with reference to FIG.
When the bit stream is input to the frame decomposition unit 102 of the decoder 101, the frame decomposition unit 102 performs an error check, separates ancillary bits of audio data, and outputs the data to the outside. The audio data other than the ancillary bit is sent to the restoration unit 103, dequantized, further subjected to an inverse filter by the inverse mapping unit 104, and
It is output as a CM audio sample. The ancillary bit output to the outside is converted into appropriate information by an external device. For example, if it is character information, it can be displayed on a display. FIG.
If there is a frame in which the audio signal is encoded following the frame described in the bit stream format, the audio signal can be decoded and output almost at the same time when the character is displayed on the display. . Further, the frame described in the bit stream format of FIG. 5 can exist at an arbitrary position in the bit stream.

(Embodiment 10) FIG. 6 shows a bit stream format of the MPEG1 audio encoding layer 1. A bit stream is composed of one or more consecutive frames. FIG. 6 describes only the format of one frame. The frame includes a header 61, audio data 62, and ancillary data 63. Audio data 62 consists only of bit allocation information. In the case of the layer 1 stereo channel, as in (Embodiment 7), the number of bits used for bit allocation is 4 bits × 32 subbands × 2 channels = 256 bits.

For example, if encoding is performed at a bit rate of 256 kbit / s, and the sampling frequency of an input audio sample is 48 kHz, the number of bits allowed in one frame of a bit stream is 2048 bits. . Since the number of bits used for the header is 32 bits, the number of bits available for ancillary data is 204 bits.
8- (32 + 256) = 1760 bits. Assuming that Chinese characters are written at 16 bits / character in the ancillary data, 110 characters have been written.

The operation of the decoder will be described with reference to FIG.
When the bit stream is input to the frame decomposition unit 102 of the decoder 101, the frame decomposition unit 102 separates audio data and ancillary data, and outputs the audio data to the restoration unit 103 and the ancillary data to the outside.
The audio data is sent to the restoration unit 103, dequantized, and subjected to an inverse filter by the inverse mapping unit 104,
It is output as a CM audio sample. The ancillary data output to the outside is converted into appropriate information by an external device. For example, if it is character information, it can be displayed on a display. FIG.
If there is a frame in which the audio signal is encoded following the frame described in the bit stream format, the audio signal can be decoded and output almost at the same time when the character is displayed on the display. . The frame described in the bit stream format of FIG. 6 can exist at an arbitrary position in the bit stream.

(Embodiment 11) FIG. 7 shows a bit stream format of the MPEG1 audio encoding layer 2. A bit stream is composed of one or more consecutive frames. FIG. 7 describes only the format of one frame. The frame includes a header 71, audio data 72, and ancillary data 73. Audio data 72 consists only of bit allocation. In the case of a layer 2 stereo channel, as in (Embodiment 2), the number of bits used for bit allocation is 176 bits.

For example, when encoding is performed at a bit rate of 256 kbit / s, when the sampling frequency of an input audio sample is 48 kHz, the number of bits allowed for one frame of a bit stream is 6144 bits. Since the number of bits used for the header is 32 bits, the number of bits usable for ancillary data is 2048−32 = 2
This is 016 bits. Assuming that Chinese characters are written in the ancillary data at 16 bits / character, 126 characters have been written.

The operation of the decoder is the same as that of the layer 10 in the tenth embodiment.

(Embodiment 12) FIG. 8 shows a bit stream format of the MPEG1 audio encoding layer 3. A bit stream is composed of one or more consecutive frames. FIG. 8 describes only the format of one frame. The frame includes a header 81 and audio data 82. The audio data 82 includes only a main information start code, a private bit, and an ancillary bit.

The case of a layer 3 stereo channel will be described. In layer 3, when the sampling frequency is 48 kHz and the bit rate is 256 kbit / s, the bit stream 1
The number of bits allowed for a frame is 6144 bits. Since the number of bits used for the header is 32 bits,
The number of bits available for audio data is 6144-
32 = 6112 bits. 9 bits for main information start code and 3 for private bits in audio data
Bit (5 bits in single channel mode)
Since it is assigned, the ancillary bit contains 6112-
(9 + 3) = 6100 bits are allocated. Assuming that kanji is written in the ancillary bit at 16 bits / character, 381 characters have been written.

The operation of the decoder will be described with reference to FIG.
When the bit stream is input to the frame decomposing unit 102 of the decoder 101, the frame decomposing unit 102 separates the ancillary bits of the audio data and outputs it to the outside. Audio data other than the ancillary bit is restored by the restoration unit 103
And inversely quantized by the inverse mapping unit 104, and output as PCM audio samples. The ancillary bit output to the outside is converted into appropriate information by an external device. For example, if it is character information, it can be displayed on a display. If a frame in which the audio signal is encoded follows the frame described in the bit stream format of FIG. 8, the audio signal can be decoded and output almost at the same time when the character is displayed on the display. become. Further, the frame described in the bit stream format in FIG. 8 can exist at an arbitrary position in the bit stream.

[0070]

As described above, according to the present invention, an audio signal encoding method in which a bit stream includes at least one frame composed of only a header, an error check, bit allocation, and ancillary data is provided. , A sufficient amount of ancillary data can be recorded or transmitted. For example, consider the case where the first few frames of a bit stream consist of frames of the present invention, and the audio signal is encoded in subsequent frames. The audio data and audio information are encoded in the audio data, and the ancillary data is encoded and recorded and transmitted as the characterized information of the audio signal and the audio information in the ancillary data. Character information written in ancillary data can be decoded before decoding audio signals and audio information. This allows
A sufficient amount of character information can be output to a display or the like almost simultaneously with an audio signal or audio information. Of course, the information written in the ancillary data may be information other than characters.

A sufficient amount of ancillary data is obtained by an audio signal encoding method characterized in that a bit stream includes at least one frame composed of only a header, bit allocation and ancillary data. Can be recorded or transmitted.
In this case, extra ancillary data can be encoded because the error check is not encoded.

Further, a sufficient amount of ancillary data is reproduced by an audio signal decoding method in which a bit stream includes at least one frame composed of only a header, an error check, bit allocation, and ancillary data. It is possible to do. For example, consider the case where the first few frames of a bit stream are comprised of frames of the present invention, and the audio signal is encoded in subsequent frames. In this case, before decoding the audio signal or audio information written in the audio data, it becomes possible to decode the audio signal or the character information related to the audio information written in the ancillary data. As a result, a sufficient amount of character information can be output to a display or the like almost simultaneously with an audio signal or audio information. Of course, the information written in the ancillary data may be information other than characters.

Furthermore, a sufficient amount of ancillary data can be reproduced by a decoding method of an audio signal in which at least one frame including only a header, bit allocation and ancillary data is included in a bit stream. It becomes possible. In this case, extra ancillary data can be decoded because the error check is not encoded.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a bit stream format of layer 1 according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing bit allocation information of layer 1 in the embodiment of the present invention.

FIG. 3 is a diagram illustrating a bit stream format of layer 2 according to the embodiment of the present invention.

FIG. 4 is a schematic diagram showing bit allocation information of layer 2 in the embodiment of the present invention.

FIG. 5 is a diagram illustrating a layer 3 bit stream format according to the embodiment of the present invention.

FIG. 6 is a diagram showing a bit stream format of layer 1 in the embodiment of the present invention.

FIG. 7 is a diagram illustrating a layer 2 bit stream format according to the embodiment of the present invention.

FIG. 8 is a diagram illustrating a bit stream format of layer 3 according to the embodiment of the present invention.

FIG. 9 is a block diagram showing a basic structure of an encoder.

FIG. 10 is a block diagram showing a basic structure of a decoder.

FIG. 11 is a conventional layer 1 bit stream format diagram.

FIG. 12 is a conventional layer 2 bit stream format diagram.

FIG. 13 is a conventional layer 3 bit stream format diagram.

[Explanation of symbols]

11, 31, 51, 61, 71, 81, 111, 12
1, 131 Header 12, 32, 52, 112, 122, 132 Error check 13, 33, 62, 72 Audio data (bit allocation) 14, 34, 63, 73, 114, 124 Ancillary data 53, 82 Audio data ( Main information start code, private bit, ancillary bit) 91 encoder 92 mapping unit 93 psychoacoustic unit 94 quantization and coding unit 95 frame assembling unit 101 decoder 102 frame decomposing unit 103 restoration unit 104 inverse mapping unit 113 audio data ( Bit allocation, scale factor, sample 123 Audio data (bit allocation, scale factor selection information, scale factor, sample) 133 Audio data (private bit, scale factor selection information, quantization step width information) Scale factors, Huffman coding data, Anne Sila ribitol, and other data)

Claims (12)

[Claims] 1. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 1 format, wherein at least a frame including only a header, an error check, a bit assignment, and an ancillary data is included in the bit stream. A method for encoding an audio signal, comprising one or more. 2. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 2 format, wherein at least a frame including only a header, an error check, a bit assignment, and an ancillary data is included in the bit stream. A method for encoding an audio signal, comprising one or more. 3. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 3 format, wherein at least one frame composed of only a header, an error check, a private bit, and an ancillary bit is included in the bit stream. A method for encoding an audio signal, wherein at least one is included. 4. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 1 format, wherein at least one frame composed of only a header, bit allocation and ancillary data is included in the bit stream. An audio signal encoding method characterized by being included. 5. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 2 format, wherein at least one frame composed of only a header, bit allocation and ancillary data is included in the bit stream. An audio signal encoding method characterized by being included. 6. An encoding method of an audio signal for generating an audio bit stream according to an MPEG1 audio layer 3 format, wherein the bit stream includes at least one frame composed of only a header, a private bit, and an ancillary bit. An encoding method of an audio signal, characterized by being performed. 7. A decoding method of an audio signal for reproducing an audio bit stream according to an MPEG1 audio layer 1 format, wherein at least one frame including only a header, an error check, a bit allocation, and ancillary data is included in the bit stream. A method for decoding an audio signal, wherein at least one is included. 8. A decoding method of an audio signal for reproducing an audio bit stream according to an MPEG1 audio layer 2 format, wherein at least one frame including only a header, an error check, bit allocation, and ancillary data is included in the bit stream. A method for decoding an audio signal, wherein at least one is included. 9. A decoding method of an audio signal for reproducing an audio bit stream according to an MPEG1 audio layer 3 format, wherein at least one frame composed of only a header, an error check, a private bit, and an ancillary bit is included in the bit stream. A method for decoding an audio signal, which is included in the above. 10. A decoding method of an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 1 format, wherein the bit stream includes at least one frame composed of only a header, bit allocation, and ancillary data. A method for decoding an audio signal. 11. A decoding method of an audio signal for reproducing an audio bit stream according to an MPEG1 audio layer 2 format, wherein the bit stream includes at least one frame composed of only a header, bit allocation, and ancillary data. A method for decoding an audio signal. 12. A decoding method of an audio signal for reproducing an audio bit stream according to the MPEG1 audio layer 3 format, wherein the bit stream includes at least one frame composed of only a header, a private bit, and an ancillary bit. A method for decoding an audio signal.