JPH09198607A - Voice reproducing circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reproduce a voice by running a magnetic tape faster than the speed at the time of an ordinary reproducing by providing a selection circuit, an interpolation circuit, a speech speed conversion circuit, etc. SOLUTION: This voice reproducing circuit is applied to such a magnetic tape that the same voice data are repeated recorded in respective consecutive plural blocks in the unit of blocks each consisting of plural lines of tracks and also the voice data are updated for every plural blocks. An error information circuit 1 detects error information by inputting reproduced data read out from the plural blocks each having the same voice data recorded. A selector 7 selects normal voice data based on the error information and an interpolation circuit 8 applies an interpolation processing to selected voice data to interpolate voice data which are impossible to be reproduces as a voice due to errors. A speech speed conversion circuit 10 performs a time base conversion for reproducing a voice in the same musical interval as that at the time of the ordinary reproducing with respect to the voice data applied with the interpolation processing.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio reproducing circuit for reproducing audio from a magnetic tape on which digital audio data is recorded.

[0002]

2. Description of the Related Art HD (High Definition) Digital VCR
At the conference, a digital VTR for recording and reproducing a video signal and an audio signal as digital data is standardized. Here, when recording a signal on a magnetic tape,
As shown in FIG. 10, a plurality of tracks are repeatedly formed by helical scanning, and digital video data and digital audio data within a period corresponding to one frame of a video signal are recorded over 10 tracks in the NTSC specification. It will be.

An audio data recording area S is formed at the beginning of each track, and in the 2-channel mode,
In the audio data recording area S of 5 tracks in the first half of 10 tracks, 1 frame of left channel audio data,
In the latter half of the 5-track audio data recording area S, one frame of right channel audio data is recorded. still,
To the audio data recorded in one audio data recording area, a sequence number for identifying a frame, a track pair number for identifying a track, a sync block number, and a parity are added as ID data.
Shuffling is performed for each frame in units of 6-bit data and then recorded in each audio data recording area. Image-compressed video data is recorded in the video data recording area following the audio data recording area S of each track.

By the way, in the above-mentioned digital VTR, the magnetic tape is run at a higher speed than during normal reproduction.
Special playback such as fast-forward playback can be performed. For example, in a 180-degree opposite two-head type VTR, two azimuth tracks A and B are repeatedly formed as shown in FIG. 10. When performing double speed reproduction, the magnetic tape is reproduced at twice the normal speed. By running, both heads scan two adjacent tracks diagonally,
Signal reproduction by two azimuths A and B is alternately performed.
As a result, the audio data recording areas (12a) (12b) (12c) (12d) (12) where the azimuth during recording and the azimuth during playback match
The audio data will be read only from e).

Here, the audio data recording areas (12a) (12b) (1
The data read from 2d) is the left-channel audio data, and the data read from the audio-data recording areas (12c) and (12e) is the right-channel audio data. Further, since the audio data recorded in these areas is shuffled, there is no continuity on the time axis in the read order. Therefore, at the time of audio reproduction, it is necessary to perform a process of returning the read data to the original time series data for each channel.

[0006]

However, during the special reproduction, only a part of the audio data of each frame is reproduced as shown in FIG. 10, and the audio data in the area shown by hatching in the figure is reproduced. Will be missing. Therefore, even if the process to restore the original time series data is performed,
Normal data cannot be reproduced due to the lack of data. Therefore, conventionally, a method has been adopted in which only video is displayed during special reproduction, and audio is not output by mute processing.

Therefore, an object of the present invention is to provide an audio reproducing circuit capable of reproducing audio even during special reproduction.

[0008]

According to the present invention, when recording a signal on a magnetic tape, the same audio data is repeatedly recorded in each of a plurality of consecutive blocks in units of a block composed of a plurality of tracks. , The audio data is updated for each of the plurality of blocks. Then, during special reproduction, the magnetic tape is made to run at a higher speed than during normal reproduction, helical scanning is performed, and sound is reproduced.

In the audio reproducing circuit according to the present invention, audio data sequentially read from a plurality of blocks in which the same audio data is recorded is input in parallel to the selecting means for each block, and an error contained in the audio data is included. One normal block of voice data is selected based on the information, the voice data output from the selecting means is input to the interpolating means, and the voice data that cannot be reproduced as voice due to an error is supplemented by interpolation, This is to generate normal voice data.

In the audio reproduction circuit, the same audio data is recorded in a plurality of blocks, and when an error occurs in the process of reading the audio data from each of these blocks, the error occurs randomly for each block. To do. Therefore, even if an error occurs in reading the audio data from one block, the data may be normally read from any of the other blocks.

Therefore, by selecting the sound data of one normal block based on the error information included in the sound data, a series of sound data with less error data can be obtained. The voice data that cannot be reproduced as voice due to an error is supplemented by interpolation processing using normal voice data before and after the voice data. As a result, a series of audio data similar to the original audio data recorded on the magnetic tape can be obtained.

Further, the audio reproducing circuit according to the present invention receives audio data read from a plurality of blocks in which the same audio data is recorded as an input, and based on error information contained in the audio data, one normal audio data is read. Selection means for selecting the voice data of the block, and an interpolation means for performing an interpolation process on the selected voice data to compensate for data that cannot be reproduced as voice due to an error, and for the voice data subjected to the interpolation process, Data processing means for performing data processing for reproducing sound with the same pitch as during normal reproduction.

In the audio reproducing circuit, the audio data of one normal block is selected from the audio data of each block at the same time and is output as the audio data of the time. If there is no normal voice data, for example, information indicating that an error has occurred is output. As a result, a series of audio data with little error data can be obtained. After that, voice data that cannot be reproduced as voice due to an error is supplemented by interpolation processing using normal voice data before and after the voice data.

However, the interpolation process increases the amount of data per unit time to be processed when reproducing the voice, and the voice cannot be reproduced in synchronism with the running of the magnetic tape. Decimate the audio data by a percentage. In addition, necessary data processing is performed in order to avoid changing the pitch due to thinning of the audio data.

For example, a magnetic tape is played N times (N
Is a random number), the data processing means samples a series of interpolated audio data in a cycle corresponding to the tape running speed in order to reduce the data amount to 1 / N. The audio data thus obtained is subjected to a process of expanding the time axis N times.

In this specific configuration, the audio data is thinned out by the sampling so that the audio can be reproduced in synchronization with the running of the magnetic tape, and the change of the pitch can be avoided by the conversion of the time axis. .

Further, in another specific configuration, the data processing means extracts a series of audio data extracted in a period of 1 / N of a predetermined constant cycle for each constant cycle,
This is to connect the extracted audio data.

In the specific configuration, the audio data is extracted at regular intervals to thin out the audio data, and to avoid a change in pitch of the audio data within the constant cycle. Further, by connecting the extracted audio data, it is possible to reproduce the audio in synchronization with the running of the magnetic tape.

However, since data discontinuity occurs at the concatenation part of the audio data, it is desirable to filter the data of the concatenation part to smooth the change of the data.

[0020]

According to the sound reproducing circuit of the present invention, sound can be reproduced even when special reproduction is carried out by running the magnetic tape at a higher speed than during normal reproduction.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment in which the present invention is applied to a digital VTR having two 180-degree opposed heads will be specifically described below with reference to the drawings.

First, the structure of the signal recording system which is the premise of the present invention will be described. As shown in FIG. 10, in the 2-channel mode in which the left channel and right channel audio can be reproduced during normal reproduction, one frame of the video signal is recorded over 10 tracks, and the audio signal is recorded. Shows the left channel data on the first 5 tracks,
Right half channel data is assigned to the latter half 5 tracks, but in the present invention, as shown in FIG. 4, the same audio data, for example, left channel data, is repeatedly recorded on the first half 5 tracks and the latter half 5 tracks. Then, the data is updated every 10 tracks.

In the 4-channel mode, normally, the audio data of A channel and B channel is recorded on the first 5 tracks, and the audio data of C channel and D channel is recorded on the latter 5 tracks. However, in the present invention, as shown in FIG. 5, the same audio data, for example, left channel and right channel data (Lch + Rch) is repeatedly recorded on the first 5 tracks and the latter 5 tracks, and data is recorded every 10 tracks. Will be updated.

In order to realize the above signal recording, as shown in FIG. 3, one of the left channel audio data and the right channel audio data is selected via the selector (28), and the selected one frame of audio is selected. The data is stored in the memory (29). The audio data for one frame stored in the memory (29) is read by the shuffle circuit (30) in a predetermined order and subjected to known shuffling. Here, the operations of the selector (28) and the shuffle circuit (30) are controlled by a mode signal for designating either the 2-channel mode or the 4-channel mode. For example, in the 2-channel mode, the selector (28) always selects the left channel audio data (Lch), and the memory (29) stores
The left channel audio data for one frame is written to a predetermined address twice in the first half and the second half of one frame period. In the 4-channel mode, the left channel audio data and the right channel audio data for one frame are respectively written to the predetermined addresses twice in the first half and the second half of the one frame period.

The data output from the shuffle circuit (30) is passed through a correction code adding circuit (31) and a sync / ID adding circuit (32) to add necessary code data, and then the mixing circuit (3
In 3), the video data and subcode are mixed. The data output from the mixing circuit (33) is supplied to the magnetic head after being subjected to necessary modulation processing such as 24-25 modulation through the modulation circuit (34) and then written on the magnetic tape. As a result, the track pattern shown in FIG. 4 is formed in the 2-channel mode, and the track pattern shown in FIG. 5 is formed in the 4-channel mode.

FIG. 1 shows the configuration of a signal reproducing system for the 2-channel mode. The reproduction data read from the magnetic tape goes through a well-known sync protection circuit and error correction circuit (both not shown) to compensate for the missing sync signal and correct the reproduction data error. 1
Is input to the signal reproduction system.

Here, the reproduction data includes audio data, video data, code data, etc., and is sent in units of one sync block. At the beginning of each sync block, as the ID data, a sequence number for identifying a frame, a track pair number for identifying a track, a sync block number for identifying a sync block, and these data Parity for error detection is added. Further, error information indicating that fact is added to a sync block that cannot be error-corrected beyond the error correction capability of the error correction circuit.

The reproduced data is first of all an error information detection circuit.
Input to (1), the presence or absence of error information is detected, the sync block to which the error information is added is discarded, and only the normal sync block that has no error or is error-corrected is extracted, Subsequent voice data extraction circuit (2)
Output to The error information detection circuit (1) determines whether the parity included in each sync block matches the parity generated from the sequence number, track pair number, and sync block number included in the sync block. If they do not match, it is determined that the sequence number, the track pair number, or the sync block number is unknown in the sync block, and error information indicating that is unknown is added to the sync block, and then the audio data extraction circuit (2 ). Voice data extraction circuit (2)
Then, based on the sync block number, it is determined whether the input sync block includes audio data or video data, and only the sync block including audio data is extracted and output to the subsequent stage.

Data for one frame extracted as a sync block including audio data is stored in the memory (3). Here, it is assumed that the audio data is written in the memory (3) at an address corresponding to the sync block number and the track pair number included in the ID data, and the sync block in which the ID data has an error cannot be reproduced. As a sync block, the audio data is discarded and is not written in the memory (3). As a result, only normal sync blocks read from the first half 5 tracks of 10 tracks corresponding to one frame are written in the memory (3), and normal sync blocks read from the latter half 5 tracks are also written. Will be written. An error code for recognizing that it is a non-reproducible sync block is written in an address where no data has been written.

The audio data for one frame written in the memory (3) is read by the deshuffling circuit (4) in the original data order having continuity with respect to the time axis and subjected to deshuffling. The deshuffled data is the first data separation circuit (5) and the second data separation circuit (6).
The first data separation circuit (5) separates the audio data read from the first half 5 tracks of 10 tracks corresponding to one frame, and the second data separation circuit (6) supplies the latter half of the audio data. The audio data read from the 5 tracks is separated, both data are arranged on the same time axis, and then supplied to the selector (7). The selector (7) selects and outputs one of the data when one of the data is normal data that does not include an error code, and when both are normal data or both of them are error codes. If it is, one of the predetermined data is selected and output.

For example, the data string shown in FIG. 6A is read from the first half 5 tracks of each frame, and the latter half 5
When the data string shown in FIG. 2B is read from the track, many data loss occurs in each data string due to an error in the audio data or ID data, as indicated by the mark X in the figure. Even if the data of one of the series is erroneous, the data of the other series may be normal.
Therefore, when either one of the audio data is normal as shown in FIG. 6C, a series of audio data with few missing parts can be obtained by selectively outputting the data.

By the way, during special reproduction at double speed, for example, the magnetic tape runs at twice the normal speed.
By scanning the frame period, the audio data for two frames is read out. Therefore, if it is assumed that the audio is reproduced using all the read audio data, a time axis shift occurs between the reproduction of the audio and the running of the magnetic tape, and the audio reproduction catches up with the running of the magnetic tape. There will be no.

Therefore, in this embodiment, the output signal of the selector (7) is supplied to the interpolation circuit (8) to thin out the audio data at a rate according to the tape running speed. For example, at double speed, the sound data after deshuffling shown in FIG.
As shown in (b), sampling is performed again at a cycle twice the sampling cycle. When the speed is 1.5 times, the deshuffled audio data shown in FIG. 8A is sampled again at a period 1.5 times the sampling period as shown in FIG. 8B. At this time, if the original data to be sampled is missing, the missing data is interpolated using normal data before and after, as indicated by the square marks in FIGS. 7 (a) and 8 (a). . As the interpolation method, an averaging process, a Lagrange interpolation method, a Newton interpolation method, or the like can be adopted.

The interpolation circuit (8) shown in FIG. 1 includes an interpolation buffer for temporarily storing data during the interpolation process.
(9) is connected. The interpolation circuit (8) needs to output the data after the interpolation processing at a constant time interval, whereas the time interval of the data obtained through the deshuffle circuit (4) is not constant and is still normal. Interpolation calculation cannot be performed until a predetermined number of data are collected. Therefore, the time difference between the input and output data of the interpolation circuit (8) is absorbed by the interpolation buffer (9). As a result, a series of data strings that can be recognized as voice are obtained from the interpolation circuit (8).

However, since the frequency of the voice data output from the interpolation circuit (8) increases in proportion to the multiple speed number, the time axis is proportional to the multiple speed number by the well-known speech speed conversion circuit (10). Then, it is supplied to the speaker device through the D / A converter. As a result, monaural sound is reproduced with the same pitch as during normal reproduction.

FIG. 2 shows the configuration of the signal reproducing system for the 4-channel mode. The reproduction data from the magnetic tape is first input to the error information detection circuit (11),
The presence or absence of error information is detected, the sync block to which the error information is added is discarded, and only the normal sync block that has no error or is error-corrected is extracted, and the audio data extraction circuit (12 ) Is output to. or,
The error information detection circuit (11) determines whether the parity included in each sync block matches the parity generated from the sequence number, track pair number and sync block number included in the sync block, and does not match. In this case, the error information is added to the sync block and then output to the audio data extraction circuit (12). The audio data extraction circuit (12) determines, based on the sync block number, whether the input sync block contains audio data or video data, and extracts only the sync block containing audio data. Output to the latter stage.

Data for one frame extracted as a sync block including audio data is stored in the memory (13). As in the case of 2 channels, it is assumed that audio data is written in the memory (13) at an address corresponding to the sync block number and the track pair number, and the sync block in which the ID data has an error cannot be reproduced. As a sync block, the audio data is discarded and is not written in the memory (13). As a result, in the memory (13), the audio data of the left channel and the right channel read from the first 5 tracks of 10 tracks corresponding to one frame is written, and the left channel read from the last 5 tracks is written. And the audio data of the right channel will be written. An error code for recognizing that it is a non-reproducible sync block is written in an address where no data has been written.

The left channel audio data for one frame written in the memory (13) is read by the first deshuffle circuit (14) in the order of original data having continuity with respect to the time axis, and is deshuffled. To be done. Also, the right channel audio data for one frame is sent to the second deshuffle circuit (1
By 5), the data is read in the order of the original data having continuity with respect to the time axis, and deshuffling is performed.

The left channel audio data output from the first deshuffle circuit (14) is simultaneously supplied to the first data separation circuit (16) and the second data separation circuit (17), and the first data separation circuit (16) is supplied. Then, while separating the left channel audio data read from the first half 5 tracks of 10 tracks corresponding to 1 frame, the second data separation circuit (17) separates the left channel audio data read from the latter half 5 tracks. After separating and arranging both data on the same time axis,
Supply to selector (20). The first selector (20) selects and outputs one of the data when one of the data is normal data that does not include error information, and when both are normal data or both of them are normal data. If it is an error code, one of the predetermined data is selected and output.

On the other hand, the right channel audio data output from the second deshuffle circuit (15) is the third data separation circuit (18).
And to the fourth data separation circuit (19) at the same time, and the third data separation circuit (18) separates the right channel audio data read from the first half 5 tracks of 10 tracks corresponding to one frame, The fourth data separation circuit (19) separates the right channel audio data read from the latter half five tracks, arranges both data on the same time axis, and supplies them to the second selector (21). Second selector (2
In 1), when either one of the data is normal data that does not include error information, the data is selected and output, and when both are normal data, or both are error codes. In this case, one of the predetermined data is selected and output.

The first selector (20) and the second selector (2
The output signals of 1) are the first interpolation circuit (22) and the second interpolation circuit, respectively.
(23), and the audio data after interpolation processing is thinned out at a rate according to the tape running speed. First and second interpolation circuit (22)
First and second interpolation buffers (24) and (25) for temporarily storing data during the interpolation process are connected to (23). As a result, a series of left channel audio data strings is obtained from the first interpolation circuit (22), and the second interpolation circuit (23)
From, a series of right channel audio data sequences is obtained.

The two-channel audio data output from both interpolation circuits (22) and (23) are time-axis converted by the first and second speech speed conversion circuits (26) and (27), respectively, and then D It is supplied to the speaker device via the / A converter. by this,
Stereo sound will be reproduced at the same pitch as during normal playback.

In the above embodiment, after the voice data is thinned out at a constant cycle, the change of the pitch is avoided by the speech speed conversion circuit. For example, at the double speed, as shown in FIG. 9 (a).
By thinning out the audio data of the even-numbered frames and concatenating the audio data of the odd-numbered frames with each other,
At the same time that the running of the magnetic tape and the audio reproduction are synchronized, it is possible to realize the audio reproduction with no change in pitch.

However, as shown in FIG. 9 (a), the connecting portion of the audio data becomes discontinuous, which causes noise. For example, as shown in FIG. 9 (b), the first frame and the third frame are connected. When interpolating parts, the last plural data of the first frame and the first plural data of the third frame are invalidated, and plural data before and after the invalidated data are indicated by double circles in the figure. Is used to perform interpolation processing by the Lagrange interpolation method, the Newton interpolation method, or the like, and the data generated by interpolation as shown by the square marks in the figure are replaced with the invalidated data. As a result, as shown in FIG. 9C, audio data in which the connecting portion between frames changes smoothly is obtained, and noise is prevented from occurring.

The description of the above embodiments is for the purpose of explaining the present invention, and should not be construed as limiting the invention described in the claims or reducing the scope thereof.
In addition, the configuration of each part of the present invention is not limited to the above-described embodiment, and it goes without saying that various modifications can be made within the technical scope described in the claims.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a signal reproduction system for a 2-channel mode.

FIG. 2 is a block diagram of the same for a 4-channel mode.

FIG. 3 is a block diagram showing a configuration of a signal recording system.

FIG. 4 is a diagram illustrating a track pattern in a 2-channel mode.

FIG. 5 is the same view as above in the 4-channel mode.

FIG. 6 is a diagram illustrating the operation of a selector.

FIG. 7 is a diagram for explaining the operation of the interpolation circuit at double speed.

FIG. 8 is a diagram for explaining the operation of the interpolation circuit at 1.5 × speed.

FIG. 9 is a diagram illustrating another embodiment of data thinning processing.

FIG. 10 is a diagram showing an audio data recording area of a track formed on a magnetic tape and a scanning locus during special reproduction.

[Explanation of symbols]

 (1) Error information detection circuit (2) Audio data extraction circuit (3) Memory (4) Deshuffle circuit (5) First data separation circuit (6) Second data separation circuit (7) Selector (8) Interpolation circuit (10 ) Speech rate conversion circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G11B 20/18 572 9558-5D G11B 20/18 572G 574 9558-5D 574D

Claims (5)

[Claims] 1. A magnetic tape in which the same audio data is repeatedly recorded in each of a plurality of continuous blocks in units of a block composed of a plurality of tracks, and the audio data is updated for each of the plurality of blocks. Is a circuit for reproducing audio by running the magnetic tape at a higher speed than during normal reproduction, in which audio data sequentially read from a plurality of blocks in which the same audio data is recorded Input in parallel to the selection means,
The sound data of one normal block is selected based on the error information included in the sound data, the sound data output from the selecting means is input to the interpolating means, and sound data that cannot be reproduced as sound due to an error is generated. An audio reproducing circuit characterized by supplementing by interpolation to generate a series of normal audio data. 2. A magnetic tape in which the same audio data is repeatedly recorded in each of a plurality of consecutive blocks in units of a block composed of a plurality of tracks, and the audio data is updated for each of the plurality of blocks. Is a circuit for reproducing audio by running the magnetic tape at a higher speed than during normal reproduction, and using the audio data read from a plurality of blocks in which the same audio data is recorded as an input, Selection means for selecting sound data of one normal block based on error information included in the sound data, and interpolation for compensating sound data that cannot be reproduced as sound due to an error by performing interpolation processing on the selected sound data. Means for performing data processing for reproducing the voice with the same pitch as in normal reproduction, for the voice data subjected to the interpolation process. An audio reproduction circuit comprising a data processing means. 3. When reproducing audio, the magnetic tape is run at a speed N times (N is an arbitrary real number) speed during normal reproduction, and the data processing means sets a series of audio data output from the interpolating means to the tape running speed. The audio reproduction circuit according to claim 2, comprising sampling means for sampling at a constant period according to the time axis, and time axis conversion means for performing processing for expanding the time axis by N times the audio data obtained by sampling. 4. The magnetic tape is made to run at a speed N times (N is an arbitrary real number) speed during normal reproduction during voice reproduction, and the data processing means is set within 1 / N period of a predetermined constant cycle. 3. The audio reproduction circuit according to claim 2, wherein the audio data output from the selecting means is extracted at each fixed cycle, and the extracted audio data is connected. 5. The audio according to claim 4, wherein the data processing means comprises a filter means for smoothing the change of the data of the connection portion when the audio data extracted at the constant cycle is connected. Reproduction circuit.