JPH04105278A - Noise eliminating method for compression music signal - Google Patents

Abstract

PURPOSE:To eliminate a noise by a bit compression data unit in which an error occurs by detecting the error at every bit compression data unit in which the error occurs, eliminating the data unit in which data is dissipated by the error, and connecting the data unit neighboring to the data unit in which the error occurs by cross-fading. CONSTITUTION:The recording of the data is performed along the recording track of a magneto-optical disk 2 by applying a modulation magnetic field in accordance with recording data with a magnetic head 4 in a state where the magneto-optical disk 2 driven rotatably with a spindle motor 1 is irradiated with a laser beam by an optical head 3. Also, the reproduction of the data is performed by tracing the recording track of the magneto-optical disk 2 with the laser beam by the optical head 3. The error occurring at every bit compression data unit is detected, and the data unit in which the data is dissipated by the error is eliminated, and the data unit neighboring to the one in which the error occurs is connected by the cross-fading. Thereby, the noise by the bit compression data unit in which the error occurs can be eliminated.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to a method for removing noise from compressed music signals, for example, bit-compressing input data obtained by digitizing a continuous signal and recording it on a disk-shaped recording medium. BACKGROUND ART For example, in a CD player that plays a compact disc (CD) having a concentric trunk in which digital audio data is recorded as a pit string, a spindle motor drives the disc at a constant linear velocity (CLV: Con5).
The digital audio data is reproduced by irradiating a laser beam along the track on the disk while rotating at a tangent linear velocity and detecting changes in the intensity of the reflected light depending on the presence or absence of the bit. ing.

In addition, in CD players, the error rate during data reproduction may be, for example, about 10-5.
Cross Interlead (CIRC) is a combination of Solomon code and a means to convert Hurst error into random error from interleaved error.
The error correction process employs the ve Reed Solomon Code), so that there is no problem in normal use environments.

Note that the above-mentioned compact disc (CD) uses EFM (Eight to
Fourteen! 24-bit synchronization signal given as '1 odulation) data, 14-bit (
One frame consists of 58 bits consisting of a subcode of 1 symbol), 14 x 32 bits (32 symbols) of data such as performance information and parity, and 3 and 7 margin bits between each symbol, and 98 bits. A data format in which frames are suffix blocks has been standardized.

Additionally, compact discs (CDs) have traditionally been stored in read-only memory (ROM: React Only Memory).
The so-called CD-1 (CD-Interactive), which simultaneously records image data, text data, etc. in addition to audio information and the standard of CD-ROM used as
In the e) method, seven modes are standardized for audio information, as shown in FIG. 10, for example.

In the above CD-■ standard, the sound quality level is the current level 1.
In the CD-DA mode equivalent to 6-bit PCM, the sampling frequency is 44.1 kHz, and the quantization number is 16 bits.
In A-level stereo mode and A-level monaural mode, which have a sound quality equivalent to that of an LP record, ADPCM (^ adaptive Delta
In the B-level stereo mode and B-level monaural mode, which have sound quality equivalent to FM broadcasting, ADPC with a sampling frequency of 37.8 kHz and a quantization number of 4 bits is used.
M is used, and C level, which has sound quality equivalent to AM broadcasting.
In the stereo mode and the C-level monaural mode, ADPCM with a sampling frequency of 18.9 kHz and a quantization number of 4 bits is used.

That is, as shown in Fig. 10, compared to CD-DA mode, in A-level stereo mode, the bit saving rate is 1/2, and data is recorded in so-called every two sectors (■ indicates data is recorded). The playback time for one disc is approximately 2 hours. In A level monaural mode, the bit loss rate is 1/4, data is recorded every 4 sectors, and the playback time is approximately 2 hours. It's been 4 hours,
In B-level stereo mode, the bit saving rate is 1/4
Therefore, data is recorded every 4 sectors, and the playback time is approximately 4 hours. In B level monaural mode, the bit saving rate is 1/8, data is recorded every 8 sectors, and the playback time is approximately 8 hours. In C level stereo mode, the bit saving rate is 1/8, data is recorded every 8 sectors, and playback time is approximately 8 hours.
In level monaural mode, bit saving rate is 1/16
Therefore, data is recorded every 16 sectors, and the playback time is approximately 16 hours.

One suffix code block of audio data in the above-mentioned CD-1 system conforms to the data format of CD-ROM, and as shown in the 11th section, a synchronization pattern (Sync) of 12 hides, a header (Head) of 4 hides, etc.
r), 8 hide subheader (Subheader)
, 18 sound groups with 128 hides each (
SGo+) ~ (SG, 8), 20 hide space (Space) and 4 byte reserve (Rese
rved). Furthermore, one sound group includes four sound units (S[Il) to (SU,), as shown in the 12th leap, and a sound unit consists of 28 sound data (9, J), Four same sound parameters (SPJ) are added.

Here, the above-mentioned subscript is a data sampling number, and the above-mentioned subscript "" is a sound unit number.

Problems to be Solved by the C0 Invention By the way, in the digital audio data playback system that employs CIRC as described above, when a burst error occurs in the playback data that exceeds the correction ability, the error is spread evenly over the interleaved ellipse. Errors are dispersed, making them less noticeable, and in CD players that handle continuous PCM audio data, interpolation can be performed using surrounding data, so it does not pose much of a problem in terms of audibility.

However, in the CD1 format, which handles compressed audio data, if a burst error occurs in the playback data that exceeds the correction ability, especially if a sound parameter is damaged, all information related to that sound parameter becomes meaningless. There is a risk that noise will be generated for a long time. For example, in the case of levels A and B of a CD-1, if one sound unit of data is lost, the loss will be 0°7.
This results in a data error of 4m5 minutes, which cannot be properly interpolated using pre-hold or linear interpolation, resulting in noise that lasts for a long time.

Therefore, in view of the above-mentioned conventional situation, the present invention compresses the bits of input data obtained by digitizing a continuous signal and records it on a disk-shaped recording medium, and reproduces the bit-compressed data from this disk-shaped recording medium. In a disk recording/playback device that obtains playback data by bit expansion, the noise generated when a large amount of consecutive data is lost in groups due to burst errors etc. can be removed. be.

03 Means for Solving the Problems The method for removing noise from a compressed music signal according to the present invention detects errors in each bit compressed data unit that occur in a recording/playback system that compresses bits of digital audio data and records/plays it back. The present invention is characterized in that a data unit in which data has been lost due to an error is removed, and data units adjacent to the data unit in which the error has occurred are connected by cross-fade.

E9 Effect In the method for removing noise from a compressed music signal according to the present invention, the noise caused by the bit compressed data unit in which the error has occurred is removed by connecting the adjacent data units of the bit compressed data unit in which the error has occurred. .

F. Embodiment FIG. 2 is a block diagram showing the configuration of a disk recording and reproducing apparatus that employs the method of removing noise from a compressed music signal according to the present invention to prevent the generation of noise.

As shown in FIG. 2, this disk recording and reproducing apparatus is equipped with a magneto-optical disk (2) as a recording medium, which is rotationally driven by a spindle motor (1), and an optical head (3) is attached to the magneto-optical disk (2). By applying a modulated magnetic field according to the recorded data using the magnetic head (4) while irradiated with laser light, the magneto-optical disk (2) is
Data is recorded along the recording trunk of the magneto-optical disk (2), and the data is reproduced by tracing the recording trunk of the magneto-optical disk (2) with a laser beam using the optical head (3).

The optical head (3) is composed of, for example, a laser light source such as a laser diode, optical components such as a collimator lens, an objective lens, a polarizing beam splitter, a cylindrical lens, and a photodetector divided into predetermined positions. It is provided at a position facing the magnetic head (4) with the disk (2) in between. When recording data on the magneto-optical disk (2), this optical head (3) is connected to a recording system head drive circuit (described later).
16), the magnetic head (4) is driven to apply a modulated magnetic field corresponding to recording data to a target track of the magneto-optical disk (2), thereby recording data by thermomagnetic recording. Also, this optical head (
3) detects a focus error by, for example, the so-called astigmatism method by detecting the reflected light of the laser beam irradiated on the target track, and detects a tracking error by, for example, the so-called bush-pull method, and When reproducing data from (2), a reproduction signal is generated by detecting the difference in the polarization angle (Kerr rotation angle) of the laser beam reflected from the target track.

The output of the optical head (3) is supplied to an RF circuit (5). This RF circuit (5) is connected to the optical head (3).
A focus error signal and a trunking error signal are extracted from the output and supplied to a servo control JB circuit (6), and the reproduction signal is binarized and supplied to a reproduction system decoder (21) to be described later.

The servo control circuit (6) includes, for example, a focus servo control circuit, a tracking servo control circuit, a spindle motor servo control circuit, and a thread servo control circuit. The focus servo control circuit controls the optical head (3) so that the focus error signal becomes zero.
) controls the focus of the optical system. Further, the trunking servo control circuit performs tracking control of the optical system of the optical head (3) so that the tracking error signal becomes zero. Further, the spindle motor servo control circuit controls the spindle motor (1) to rotate the magneto-optical disk (2) at a predetermined rotational speed. Further, the thread servo control circuit moves the optical hemometer F' (3) and the magnetic head (4) to a target track position on the magneto-optical disk (2) designated by the nostem controller (7). The servo control circuit (6), which performs such various control operations, supplies the system controller (7) with information indicating the operating status of each section controlled by the servo control circuit (6).

Further, the system controller (7) is connected to a key input operation section (8) and a display section (9).

This system controller (7) controls the recording system and the reproduction system in an operation mode specified by operation input information from the key manual operation section (8). The system controller (7) also controls the optical path (3) and magnetic It manages the recording position and playback position on the recording trunk that the head (4) is tracing.

The recording system of this disc recording and reproducing apparatus includes an A/D converter (A/D converter) to which an analog audio signal A is supplied from an input terminal (10) via a low-pass filter (11).
12).

The A/D converter (12) is configured to output the audio signal AI.
CD-D in the above-mentioned CD-1 system by quantizing N.
Digital audio data at a predetermined transfer rate (75 sectors/sec) corresponding to A mode is formed. Digital audio data obtained by this A/D converter (12) is supplied to an ADPCM encoder (13).

The ADPCM encoder (13) converts digital audio data of a predetermined transfer rate obtained by quantizing the audio signal A 1 H by the A/D converter (12) into data corresponding to various modes in the CD-1 system. It performs compression processing, and the operation mode is specified by the system controller (7). In this disc recording and reproducing apparatus, for example, as shown in FIG. 3, the digital audio data in the CD-DA mode is converted into 1
The data is compressed to /4 and the transfer speed is 18.75 (75
/4) ADP in B level stereo mode with 7 seconds sector
It is assumed that the data is converted into CM audio data.

The B-level stereo mode ADPCM audio data that is continuously output from this ADPCM encoder (13) at a transfer rate of 18.75 sectors/second is stored in the memory (
14).

Data writing and reading in the memory (14) is controlled by the system controller (7), and B-level stereo mode ADPCM audio data supplied from the ADPCM encoder (13) is stored at 18.75 sectors/sec. written continuously at a transfer rate of In addition, this memory (14) is a B-level memory that is continuously written at the transfer rate of 18.75 sectors/second.
ADPCM audio data in stereo mode is read out as recording data in bursts at the transfer rate of 75 sectors/second.

As shown in FIG. 4, the system controller (7) sets the write pointer (W) of the memory (I4) to 18
．． ADPCM audio data is continuously written to said memory (14) at a transfer rate of 18.75 sectors/second by continuously incrementing at a transfer rate of 75 sectors/second, and is stored in said memory (14). When the amount of the ADPCM audio data being stored exceeds a predetermined amount, the read pointer (
R) in a burst manner at a transfer rate of 75 sectors/second to transfer the ADPCM from the memory (14).
Memory control is performed so that only a predetermined amount of audio data is read out as recording data in a burst manner at a transfer rate of 75 sectors/second.

The ADPCM audio data, that is, the recording data read out in bursts from the memory (14) at the transfer rate of 75 sectors/second is supplied to the encoder (15).

The encoder (15) rearranges the recorded data supplied in bursts from the memory (14) in units of sound groups, performs encoding processing for error correction, EFM encoding processing, etc. CD-R
To OM! 1! For example, as shown in Figure 5, it is composed of a sound group rearrangement circuit (31), a CD-1 encoder (32), and an EFM encoder (33). Ru.

The sound group rearrangement circuit (31) of this encoder (15) consists of 17 delay circuits (DL) each having a different delay amount that is an integral multiple of the delay amount corresponding to one sector.
~(DLl,), and each sound group (SGo+
) to (SG, , ) by giving respective delay amounts to perform rearrangement processing over 18 sectors. Further, the CD-1 encoder (32) outputs the recorded data that has been rearranged in units of sound groups by the sound group rearrangement circuit (31).
Performs encoding processing for error correction in accordance with M. The EFM encoder (33) encodes the CD
-1 encoder (32) encoded output as EFM
Encode and output.

The encoder (15) operates the ADPCM audio data in the B-level stereo mode at a transfer rate of 75 sectors/second, which is faster than the normal transfer rate in the B-level stereo mode.

The recorded data encoded by the encoder (15) is sent to the magnetic head drive circuit (16).
is supplied to

This magnetic head drive circuit (16) includes the magnetic head (
4) is connected to drive the magnetic head (4) so as to apply a modulated magnetic field corresponding to the recorded data to the magneto-optical disk (2).

Further, the system controller (7) performs the above-mentioned memory control on the memory (14), and the recorded data read out in bursts from the memory (14) by this memory control is shown in FIG. The recording position is controlled so as to continuously record on the recording trunk of the magneto-optical disk (2). This recording position control is performed by managing the recording position of the recorded data read out in bursts from the memory (14) by the system controller (7), and by controlling the recording position on the recording trunk of the magneto-optical disk (2). This is done by supplying a control signal specifying the above to the servo control circuit (6).

In the recording system of this disk recording and reproducing apparatus, the memory control by the system controller (7) allows the ADPCM encoder (13) to
ADPCM output continuously at a transfer rate of sectors/second
Audio data is written to the memory (14) at the transfer rate of 18.75 sectors/second, and
) When the amount of the ADPCM audio data stored in the memory (14) exceeds a predetermined amount,
), only a predetermined amount of the ADPCM audio data is read out as recording data in a burst manner at a transfer rate of 75 sectors/second, so that the ADPCM audio data is read out in bursts at a transfer rate of 75 sectors/second, while always ensuring a data writing area of more than the predetermined amount in the memory (14). ,
Input data can be written continuously into the memory (14). The recorded data read out in bursts from the memory (I4) is controlled by the system controller (7) to control the recording position on the recording track of the magneto-optical disk (2). ) can be recorded continuously on the recording track. Moreover, as mentioned above, the memory (1
4), a data writing area of a predetermined amount or more is always secured, so the system controller (7) detects when a trunk jump or the like occurs due to a disturbance, etc., and performs a recording operation on the magneto-optical disk (2). Even if the process is interrupted, the input data can continue to be written in the data write area exceeding the predetermined amount, and the recovery process can be performed during that time, so that the input data cannot be written on the recording track of the magneto-optical disk (2). Can be recorded continuously.

The magneto-optical disk (2) has header time data corresponding to the physical address of the sector in the ADP.
It is added to the CM audio data sector by sector and recorded. In addition, catalog data indicating the recording area and recording mode is recorded at the head of the catalog.

Next, the reproduction system in this disc recording/reproducing apparatus will be explained.

This reproducing system is for reproducing recorded data continuously recorded on the recording trunk of the magneto-optical disk (2) by the recording system as described above, and is for reproducing recorded data continuously recorded on the recording trunk of the magneto-optical disk (2).
3) includes a decoder (21) to which the reproduction output obtained by tracing the recording trunk of the magneto-optical disk (2) with a laser beam is binarized and supplied by the RF circuit (5).

The decoder (21) corresponds to the encoder (15) in the recording system described above, and is, for example, the sixth decoder (21).
As shown in the figure, an E F M decoder (41) and a CD
-I decoder (42), sound group rearrangement circuit (31), and CD-■ decoder (43).

The EFM decoder (41) of this encoder (21) is
The RF circuit (5) decodes the EFM code obtained as a binarized playback output, and supplies the decoded output to the CD-1 decoder (42). This CI)-1
The decoder (42) performs decoding processing for error correction and the like. The sound group rearrangement circuit (33) has 17 delay circuits (DL, ) to
(DLI?), and by giving each delay amount to each sound group (SG,,) to (SG + s) of the playback data, rearrangement processing over 18 sectors is performed in the encoder (15) of the recording system. Return each sound group (SGOI) to (SGII) to their original state.

Note that this decoder (21) operates the ADPCM audio data in the B-level stereo mode at a transfer rate of 75 sectors/second, which is faster than the normal transfer rate in the B-level stereo mode.

The reproduced data obtained by the decoder (21) is then supplied to the memory (22).

Writing and reading of data in the memory (22) is controlled by the system controller (7), and playback data supplied from the decoder (21) at a transfer rate of 75 sectors/second is transferred at a rate of 75 sectors/second. Written in bursts at high speed. Also, this memory (22)
In this case, the playback data written in bursts at the transfer rate of 75 sectors/second is continuously read out at the normal transfer rate of 18.75 sectors/second in B-level stereo mode.

As shown in FIG. 7, the system controller (7) sets the write pointer (-) of the memory (22) to 75.
The playback data is written in the memory (22) at a transfer rate of 75 sectors/second, and the read pointer (R) of the memory (22) is incremented at a transfer rate of 18.75 sectors/second. The playback data is continuously read out from the memory (22) at a transfer rate of 18.75 sectors/sec, and the write pointer side) is transferred to the read pointer (R). When it catches up, the writing stops, and when the amount of the reproduced data stored in the memory (22) becomes less than a predetermined amount, the write pointer (-) of the memory (22) is set to 75 sectors/75 sectors. Memory control is performed by incrementing in bursts at a transfer rate of seconds.

Further, the system controller (7) performs the above-described memory control on the memory (22), and through this memory control, the playback data written in bursts from the memory (22) is transferred to the magneto-optical disk (22). The playback position is controlled so that playback is performed continuously from the recording track ). This control of the playback position is carried out by managing the playback position of the playback data read out in bursts from the memory (22) by the system controller (7), and controlling the playback position on the recording track of the magneto-optical disk (2). The control signal specifying the above servo control circuit (
6).

B obtained as playback data read out continuously from the memory (22) at a transfer rate of 18.75 sectors/sec.
ADPCM audio data in level stereo mode is supplied to an ADPCM decoder (23).

This ADPCM decoder (23) is an AD of the recording system.
It corresponds to the PCM decoder (13), and the operation mode is specified by the system controller (7).
This disk recording and reproducing apparatus expands ADPCM audio data in B-level stereo mode by four times and reproduces digital audio data in CD-DA mode. Digital audio data is supplied by this ADPCM decoder (23) to a D/A converter (24).

The D/A converter (24) converts the digital audio data supplied from the ADPCM decoder (23) into an analog audio signal A. , form T. Analog audio signal A obtained by this D/A converter (24). u7 is output from the output terminal (26) via the low-pass filter (25).

The playback system of the disc recording/playback apparatus of this embodiment also has a digital output function, and the ADPCM decoder (23) converts the digital audio data into a digital audio signal via the digital output encoder (27). Digital output terminal (28
).

The playback system in this disk recording/playback apparatus reproduces B-level stereo mode ADPCM audio data at 75 sectors/sec, which is played back from the recording track of the magneto-optical disk (2), under the memory control by the system controller (7). The ADPCM audio data is written to the memory (22) in bursts at a transfer rate of 7.
Since the data is read out continuously at a transfer rate of 5 sectors/second, the playback data is read out continuously from the memory (22) while always ensuring a predetermined amount or more of data readout area in the memory (22). be able to. Also,
The playback data read out in bursts from the memory (22) is controlled by the system controller (7) by controlling the playback position on the recording track of the magneto-optical disk (2). It can be played back continuously from the recorded track. Moreover, as mentioned above, since the memory (22) always has a data readout area of a predetermined amount or more, the system controller (7) detects when a track jump or the like occurs due to a disturbance, etc. Even when the playback operation for the magneto-optical disk (2) is interrupted, playback data can be read from the data readout area of the predetermined amount or more and output of the analog audio signal can be continued, and a recovery processing operation can be performed in the meantime. I can do it.

In a disc recording/playback device having such a configuration, as described above, each sound group (SG Ol ) of the compressed recorded data is processed in the recording system encoder (15).
Since we are performing rearrangement processing over 18 sectors for ~(SG l @), for example, if all the data in one sector (k) is lost due to a burst error, as shown with diagonal lines in Figure 8, then , the above playback decoder (2
In the reproduced data obtained through 1), as shown by diagonal lines in FIG. SGo+) to (SG+*) are distributed into one sound group.

Therefore, in the method of the present invention, if the sound unit L4 of Lc has an error as shown in FIG. 1A, the above sound unit 4 is found and cut, and Next, the data of sound units L3 and L5, which are located on both sides of sound unit 4 described above, are connected by cross-fading. At this time, the R4 sound unit is cut so that the phases of L-cM and RCH do not shift, and the data of the R3 and R5 sound units are connected by cross-fading.

If the crossfade described above is performed linearly within one sound unit, the value of each sample will be shorter by one sound unit7), for example, so there will be a total of two sound units including the cut sound unit. The flow of the music will be interrupted, but if this is the case, it will not be a problem for the auditory sense.

C1 Effect of the Invention As described above, in the compressed music signal noise removal method according to the present invention, by connecting adjacent data units of the bit compressed data unit in which the error has occurred by cross-fading, the bit compressed data in which the error has occurred is removed. Noise caused by the unit can be removed.

[Brief explanation of drawings]

FIG. 1 shows bit compressed data k in which an error has occurred by the noise removal method for compressed music signals according to the present invention: O~2
It can be set to 0. In addition, FIG. 3 is a block diagram showing the configuration of a disc recording and reproducing apparatus that employs a method of cross-fading one sound unit and removing noise from the music signal to prevent the generation of noise. FIG. 4 is a diagram showing the state of the memory under memory control in the recording system of the disc recording and reproducing apparatus, and FIG. 5 is a diagram showing the state of the encoder provided in the recording system of the disc recording and reproducing apparatus. Block diagram showing specific configuration, No. 6
FIG. 7 is a block diagram showing a specific configuration of a decoder provided in the playback system of the disk recording/playback device, FIG. 7 is a diagram showing the state of memory under memory control in the playback system of the disk recording/playback device, and FIG. The figure shows a data string when a burst error over one sector occurs in the disk recording/reproducing device, and FIG. 9 shows the state of reproduced data for the data string including the burst error shown in FIG. 8 above. , FIG. 10 is a diagram showing the data format of the CD-1, FIG. 11 is a diagram showing the audio data format of the CD-T, and FIG. 12 is a diagram showing the audio data format of the CD-1.
FIG. 3 is a diagram showing the structure of a sound group of audio data in FIG. (1)... Spindle motor (2)... (3)... (4)... (6)... (7)... (10) ..... (12) ..... (13) ..... (14) ..... (15) .... (16) ..... (21 )... (22)... (23)... (24)... (26)... Magneto-optical disk optical head magnetic head servo control circuit system controller Input terminal A/D converter ADPCM encoder memory decoder Magnetic head drive circuit decoder memory ADPCM decoder D/A converter output terminal

Claims (1)

[Claims] Detects errors in each bit-compressed data unit that occur in a recording/playback system that records and plays back bit-compressed digital audio data, and removes data units in which data is lost due to the error. A method for removing noise from a compressed music signal, characterized in that adjacent data units of a data unit in which a noise is generated are connected by cross-fade.