JP2000105600A - Recording regenerating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To solve the problem of poor sound quality and effectively utilize a memory to improve the function of a device by providing a circuit for processing the output of an A/D converter as the information not subjected to compression by a circuit of another system simultaneously with the output to a digital signal processor. SOLUTION: When a recording is instructed to a control circuit M0 by a key input means K1, voice information is inputted from an analog input circuit A1, and 16-bit digital information is outputted to a voice compressing and extending circuit D1 every 48 kHz by an A/D converting circuit A3. In this recording regenerating machine, the 16-bit digital information is also transmitted to a sampling converting circuit D2. The sampling converting circuit D2 performs, for example, the processing of curtailing it from 48 kHz to 8 kHz. Accordingly, in recording, compressed information and non-compressed low sampling information are simultaneously inputted to the control circuit M0. The control circuit M0 sequentially writes the data to memory chips M1-Mn.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information recording / reproducing apparatus.

[0002]

2. Description of the Related Art In recent years, various devices for recording, reproducing, and converting analog information such as music and movies into digital information have been developed. In addition, many devices have been developed for recording information by further compressing the information for effective use of the recording medium, and for expanding and reproducing the information.

[0003] Many information compression techniques have also been standardized or proposed, and in many devices, a digital signal processor (hereinafter referred to as DSP) incorporated in the device performs arithmetic processing.

In general, compression and decompression algorithms use the characteristics of human hearing to analyze unnecessary patterns by analyzing patterns and frequencies based on digitized analog information per unit time. Information is compressed by deleting it.

[0005] Recently, tape-based devices, disk media-based portable audio devices, and video devices using these compression techniques have been put to practical use.

A recording / reproducing apparatus in which these compression techniques are applied to a semiconductor memory is also being developed, and a large-capacity semiconductor memory has been attracting attention as an information recording medium with miniaturization and power saving of equipment.

[0007] Above all, a flash memory which is a batch erasing type memory has attracted attention as an information recording medium because a backup power supply is unnecessary and a large capacity is possible.

On the other hand, from a functional point of view, the sound is reproduced early to find the reproduction start position or the recording start position (hereinafter referred to as "early reproduction"), and is intentionally reproduced late for accurate positioning. (Hereinafter referred to as “slow playback”), and various other convenient functions for the user are also provided.

[0009]

As described above, most devices have a function of reproducing a fast-listening sound or a slow-listening sound.
In a device for compressing and recording audio, the following problem occurs during variable speed reproduction.

First, problems at the time of early listening will be described.

In the audio compression, for example, as shown in FIG. 3, information included in a fixed time interval (10 msec in FIG. 3) is compressed by a DSP and recorded.

On the other hand, in the case of normal reproduction, on the contrary, it is sufficient to read out the data in order, expand the data by the DSP, and reproduce the waveform with the same lapse of time as during recording. Processing for shortening is required.

[0013] In principle, if you try to hear twice as fast, DS times faster than 1/2 the real time during recording.
It is necessary to complete the extension processing of P.

Actually, since there is also a shortening correction in the direction of the time axis and an accompanying process on the system, it is necessary to perform the decompression process in about 1/4 of the actual time at the time of recording.

However, the processing speed of the DSP has a performance limit of the processor itself. Even if it is within the performance limit of the processor itself, the higher the processing speed, the greater the current consumption, which is a fatal drawback, especially in portable devices using batteries.

For this reason, in many portable devices that reproduce compressed information at the present time, early listening processing is performed by thinning out certain time intervals.

[0017] Therefore, the sound included in the delimited portion of the decimated time is discarded and is not reproduced, so that the continuity of the sound is impaired at the time of early listening, and the faster the early listening, the harder to hear the reproduced sound. Problem.

Next, the problem at the time of late listening will be described.

On the other hand, at the time of slow listening, the processing speed of the DSP is more advantageous than at the time of early listening, so that there is no problem in the performance of the processor itself.

However, a process of extending the passage of time is necessary, and for that purpose, the same fixed time segment voice is repeated several times.

When performing slow listening reproduction using only the waveform obtained by expanding the compressed information as described above, there is a problem that the sound loses continuity and reproduction becomes difficult to hear.

Next, the problem regarding the effective use of the memory will be described.

On the other hand, there is also a problem regarding the use efficiency and the use method of the memory. Many disk-based recording media have a fixed recording unit represented by a sector due to a physical limitation of a track.

However, the information compressed as shown in the example of the fixed recording unit of 512 bytes in FIG. 3 does not always match the number of recordable bytes of the fixed recording unit.

[0025] The portion that has thus been left is often left unused. The reason is as follows.

When compressing and recording information having a time axis such as audio information or moving image information, it cannot be realized without real-time processing. The location search method is complicated and the processing is limited and unrealistic.

The other is that it takes more time than usual for buffer flushing (physical writing from the buffer to the recording medium when the number of bytes of the fixed recording unit is accumulated). Due to time constraints, it is necessary to increase the buffer size to absorb the time.

Further, even when performing processing such as partial rewriting, if a compressed information unit is written across tracks and sectors, the processing becomes extremely complicated and processing efficiency is low, and the like. No.

In order to perform such complicated processing, a sophisticated control circuit and a buffer such as an SRAM are required. However, this leads to increased costs, so depending on the distribution of main information, it is more advantageous to use inexpensive large-capacity memory even if there is a little unused part in terms of cost, reliability and function. is there.

The same can be said for a recording medium using a semiconductor memory emulating a disk-based recording medium.

Actually, some types of flash memories have the same physical writing specification of the chip as the recording unit of the disk system.

In a sense, when considering compatibility with information recorded on a disk system, there is also an aspect that it is better to keep the same when recording on a semiconductor.

As described above, regardless of the recording medium, at present, the surplus portion is not used very effectively.

The present invention has been made in view of these problems, and aims to simultaneously solve the problem of unsightly sound and the effective use of memory, and to improve the function of the apparatus by effectively utilizing the problem.

[0035]

According to the present invention, the following functions are incorporated for the purpose of solving the above-mentioned problems.

According to the first aspect of the invention, in order to compress and record analog data, it is necessary to first convert analog information into digital information. This can be realized by an A / D converter. Further, the information is compressed by the DSP.
Further, a circuit is provided for inputting the output of the A / D converter to the DSP and processing the information as information that is not subjected to compression by another circuit.

According to a second aspect of the present invention, in the device of the first aspect, the output information of the A / D converter acquired at the sampling rate of the analog information to be compressed is determined based on a sampling rate of another system which is lower than that of the analog information. A thinning circuit.

According to the invention of claim 3, claim 1 or 2
In this apparatus, a circuit for recording the information obtained by simultaneously compressing the analog data and the information thinned out without performing the compression is provided from a circuit of another system.

According to a fourth aspect of the present invention, in the apparatus of the first to third aspects, the compressed information and the information thinned out without compression are simultaneously recorded in the same recording unit of the recording medium. A circuit is provided to write this in a buffer of the same size as that described above, and to write this when the compression information is output.

According to a fifth aspect of the present invention, in the apparatus of the first to fourth aspects, a circuit for reading only the thinned information without performing compression is provided, and a D / A circuit for reproducing at a sampling rate at the time of recording is provided. .

According to a sixth aspect of the present invention, in the apparatus of the first to fifth aspects, a circuit is provided for varying a D / A reference clock when reproducing based on thinned information without performing compression. Performs variable speed playback processing such as reviews and cues.

According to a seventh aspect of the present invention, in the apparatus of the first to fifth aspects, when performing a review, a cue, or the like,
Performs variable-speed playback without changing the pitch.

According to an eighth aspect of the present invention, in the apparatus of the first to seventh aspects, a variable speed reproduction processing algorithm which does not change the pitch based on information sampled at the time of recording processing at a low sampling rate is applied. The bit rate is further reduced by detecting a repetition pattern and recording the number of repetitions.

According to a ninth aspect of the present invention, in the apparatus of the first to eighth aspects, a circuit is provided which simultaneously performs reproduction at a low sampling rate and compression and recording operations at a high sampling rate in different systems.

According to the tenth aspect, in the first aspect,
In any of the devices (1) to (9), there is provided means for detecting a break in a phrase by applying a detection circuit for a phoneme repetition pattern.

According to the eleventh aspect, in the first aspect,
In each of the devices (10) to (10), an interface circuit is provided for separately managing the thinned information and the compressed information without performing compression and transferring the information to another device.

According to the twelfth aspect of the present invention, the first aspect of the present invention is provided.
In the apparatuses of Nos. 1 to 11, writing and reading are performed on the recording medium by an algorithm according to the writing and reading control of a fixed-length recording unit write type semiconductor memory represented by a flash memory. Alternatively, a removable card slot using a memory card and a card interface circuit are provided.

The operation of the present invention is as follows.

The operation according to the first aspect of the present invention is that processing using compressed information and uncompressed information becomes possible.

According to the second aspect of the present invention, the processing using the compressed information and the non-compressed information can be performed, and the information amount of the non-compressed information can be reduced.

An effect of the invention of claim 3 is that compressed information and non-compressed information can be simultaneously processed and can be recorded.

According to the fourth aspect of the present invention, the compressed information and the non-compressed information can be simultaneously processed and can be recorded in the same area, so that the area can be used effectively and at high speed. What you can do.

The operation according to the fifth aspect of the present invention is that the reproduction can be performed without operating the decompression circuit of the DSP since the reproduction is performed based on the non-compressed information.

An effect of the invention of claim 6 is that uncompressed information can be used for variable speed reproduction processing.

An effect of the invention of claim 7 is that uncompressed information can be used for variable speed reproduction processing and a low bit rate can be achieved.

The operation according to the eighth aspect of the present invention is that reproduction can be performed during recording even when the DSP is operating in the recording mode.

The effect of the ninth aspect is that meaningful (audible) reproduction can be performed in the reverse direction.

The operation according to the tenth aspect of the present invention is that only uncompressed information or only compressed information can be transferred or received to another device.

The operation according to the eleventh aspect of the present invention is to provide a high-performance portable semiconductor recording / reproducing apparatus utilizing a free area effectively by using a writing method of a flash memory.

[0060]

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

First, the case of recording according to the embodiment will be described. The basic configuration of the present invention will be described with reference to the block diagram of FIG. Here, a music recording / reproducing apparatus will be described as an example, with a sampling rate of 48 KHz and a resolution of 16 bits.

When recording is instructed to the control circuit (M0) by the key input means (K1), the control circuit (M0) executes the DSP operation.
Is set to the compression mode, and the A / D conversion circuit (A3) is activated.

As a result, audio information is input from the analog input circuit (A1), and the A / D conversion circuit (A3)
16-bit digital information is output every 48 KHz.

In the prior art, here, the 16-bit digital information is transmitted only to the audio compression / expansion circuit (D1) and the compressed data is read by the system control circuit (M0). , 16-bit digital information is also sent to the sampling conversion circuit (D2) at the same time.

The sampling conversion circuit (D2) performs, for example, a process of thinning this from 48 KHz to 8 KHz.

Further, the 16-bit information can be further converted into 8-bit data according to the μ-low rule utilizing human auditory characteristics. This may be performed by hardware or software in the control circuit (M0).

Thus, during recording, the control circuit (M
In (0), compressed information and uncompressed low-sampling information are input simultaneously.

Next, the control circuit (M0) sequentially writes these data in the memory chip group (M1 to Mn).

In this example, the compressed information is 424 bytes (L channel = 212 bytes / R channel = 212 bytes)
(Byte) The description will be made assuming that the uncompressed information is 80 bytes.

When writing, compressed information is recorded in the main area of FIG.

Using 84 bytes of the unused area in FIG. 2,
Record 80 bytes of uncompressed information.

The page shown in FIG. 2 is an example in which a page write type flash memory is used as the semiconductor memory.

This page corresponds to a fixed recording unit represented by a sector in many disk recording media.

Of course, the main compressed information and the non-compressed information may be recorded according to other rules and rules.

Next, normal reproduction in the embodiment will be described.

When reproduction is instructed to the control circuit (M0) by the key input means (K1), the control circuit (M0) executes the DSP operation.
The audio compression / expansion circuit (D1) is set to the expansion mode, and the D / A conversion circuit (A4) is activated.

Next, the compressed information written in the main area (shown in FIG. 2) of the memory chip group (M1 to Mn) is read by the control circuit (M0).

The audio compression / expansion circuit (D1) periodically requests compression information from the control circuit (M0), and performs expansion processing when 212 bytes (L or R channel) have been received in the example of FIG.

The expanded data is sent to the D / A conversion circuit (A4) according to the sampling rate (48 KHz in the example of FIG. 3) at the time of recording, and is reproduced from the analog output circuit (A2).

Next, reproduction of uncompressed information in the embodiment will be described.

The quick-hearing function is a function for moving the position while checking the voice to the place where the user wants to find the beginning. For example, a rehearsal for finding the beginning of a song or a phrase or specifying the division position when dividing the song is desired. This function is used when the user wants to hear the recording requirements earlier than the actual recording time.

However, as described in the section of the problem to be solved by the invention, according to the conventional method, the continuity of the sound is lost because the fast listening process is performed by thinning out the fixed time intervals. ing.

Thus, in the present embodiment, an example of fast listening using uncompressed information will be described to solve this.

The non-compressed information is recorded simultaneously with the compressed information as described in the normal recording embodiment.

First, when the key input means (K1) instructs the control circuit (M0) to perform fast listening reproduction, the control circuit (M0)
0) causes the audio compression / expansion circuit (D1), which is a DSP, to be stopped by a switching signal.

This is because decompression is not required for fast listening reproduction because low-sampling-rate uncompressed information is used.

However, the sampling conversion circuit (D3) and D
The / A conversion circuit (A4) is activated.

Next, the non-compressed information written in the main area (shown in FIG. 2) of the memory chip group (M1 to Mn) is read by the control circuit (M0).

The sampling conversion circuit (D2) periodically requests information from the control circuit (M0), is sent to the D / A conversion circuit (A4), and is reproduced from the analog output circuit (A2).

Next, a description will be given of fast listening reproduction of non-compressed information in the embodiment.

The processing flow is basically the same as that of the embodiment for reproducing the non-compressed information.

The difference is that since the waveform in the time axis direction is shortened by the sampling conversion circuit (D3), the pitch becomes higher, but the same effect as the so-called "cue reproduction" in which the tape is reproduced while the tape is being fast-forwarded by the tape recorder is provided. can get.

Next, a first example of fast listening reproduction of uncompressed information without changing the pitch in the embodiment will be described.

Basically, the processing flow is the same as that of the embodiment for reproducing the uncompressed information at a high speed.

The difference is that the waveform in the time axis direction is deleted in the sampling conversion circuit (D3), so that a so-called "variable speed reproduction" for fast-forward reproduction without changing the pitch is obtained.

More specifically, the reproduced voice waveform is stored in a memory, and sampling and rejection are alternately repeated for each basic unit, and the sampled voice waveform is stored for a time corresponding to the time length of the rejection period. The pitch is lowered by expanding the sound, thereby making it easier for the listener to hear.

An example of this type of technique is disclosed in Japanese Patent No. 2624826.

Next, a description will be given of a second example of fast-listening playback of uncompressed information without changing the pitch in the embodiment.

Basically, the flow of the processing is the same as that of the first embodiment for reproducing the uncompressed information at a high speed.

The difference is that a technique of analyzing the pattern of the waveform information to be reproduced in the time axis direction and deleting portions where the same waveform is repeated to reduce the time is applied.

If this is applied, it is possible to reproduce the sound at high speed without changing the pitch with the generated sound more natural. In the case of the present embodiment, since uncompressed waveform information has already been recorded, a waveform that is a source of analysis can be obtained without expanding compressed information.

This analysis may be carried out by a voice compression / decompression circuit (D1) DSP according to the accuracy, or its function may be incorporated in a sampling conversion circuit. Control circuit (M
This can be performed by the software processing of 0).

In order to make the sound more natural,
Since it is necessary to analyze phoneme repetition with a time width of a plurality of frames, uncompressed information of a plurality of frames is read at a time in this case.

Next, a description will be given of the slow listening reproduction in the embodiment.

The same can be said of the first and second embodiments for reproducing the uncompressed information at a high speed.

The difference is that the waveform information to be reproduced is not analyzed by repeating the pattern in the time axis direction to delete a portion where the same waveform is repeated, but rather by repeating the waveform information several times more than in real time. Is to apply.

With this, it is possible to reproduce a slow-listening sound with a more natural sound without changing the pitch.

Next, the reduction of the bit rate of the recording uncompressed information in the embodiment will be described.

As described in the second example of the embodiment for reproducing the uncompressed information without changing the pitch, the recorded uncompressed information is used for the rapid listening without changing the pitch. If it is decided to use it, there is a method to shorten the time by analyzing the pattern of the waveform information to be reproduced in advance in the time axis direction at the time of recording and deleting the part where the same waveform is repeated is there.

This is basically the same processing flow as in the above-described recording embodiment, except that the same waveform is input by repeating the waveform information input simultaneously with the recording in the time axis direction and analyzing the pattern. That is, there is a process of deleting a repeated portion and then recording it in a memory.

As a result, if only the remaining waveforms are recorded, the processing at the time of recording is increased, but at the time of reproduction, it is possible to perform the quick listening reproduction without changing the pitch by a simple processing.

Next, the simultaneous recording and reproduction of the embodiment will be described.

As described in the embodiment of reproducing uncompressed information and the embodiment of fast-listening, when reproducing uncompressed information, the audio compression / expansion circuit (D1) which is a DSP operates. You don't have to.

Therefore, even during recording as in the recording embodiment, reproduction can be performed by operating the sampling conversion circuit (D3) and the D / A converter (A4).

Next, comment recording of non-compressed information according to the embodiment will be described.

In the above embodiments, the recording of the non-compressed information is performed simultaneously with the recording of the compressed information.
When recording the compressed information, the user may be allowed to set so as not to record the uncompressed information.

In this case, if, for example, a comment recording button is pressed while the compressed information is being reproduced later, the content may be recorded at the current position as uncompressed information.

The process is the reverse of the process of reproducing the uncompressed information during recording in the embodiment of simultaneous recording and reproduction.

Next, the transfer of the embodiment will be described.

As shown in FIG. 3, the low-sampling uncompressed information is located in the gap between the main information, so that it is possible to cut out and read out only this portion, for example.

Next, the card according to the embodiment will be described.

The semiconductor memory group (M1 to Mn) in FIG. 1 can be detachably detached from the control circuit (M0).

Next, a first example of mixed writing of the embodiment will be described.

FIG. 3 illustrates the concept of recording an area in a page by dividing it into two parts. However, it is of course possible to write this arrangement in a manner in which compressed information and non-compressed information are mixed.

For example, it may be determined that compressed information is written alternately at even addresses and non-compressed information is written alternately at odd addresses.

Next, a second example of mixed writing according to the embodiment will be described.

In the first example of the mixed writing mode, an example is shown in which data is mixed in the same page. However, it is of course possible to write uncompressed information using a completely different area of the memory space.

In this case, however, the unused area in the page may be used for another purpose (for storing text information, a spare buffer for music division, etc.).

[0129]

According to the present invention, both the compressed information and the non-compressed information can be recorded by using the previously unused area. Equipment can be provided.

[0130] The first effect is that the reproduction function of the non-compressed information can be provided simply. In order to reproduce data from a recording medium on which only compressed information is recorded, it is necessary to operate a decompression circuit.

However, according to the present invention, since uncompressed information is also recorded at the same time, it can be easily reproduced only by reproducing it. The decompression circuit has a much larger processing amount per time than a general control circuit, and consumes a large amount of current because the operation clock is advanced to perform the decompression processing in real time.

This can be a fatal defect for a portable device using a battery. Therefore, in order to reduce the current consumption as much as possible, the non-compressed information according to the present invention may be reproduced in a scene other than the case where reproduction is actually performed with high sound quality. Examples of applications include the following.

The second effect is that a method of reproducing uncompressed information can be considered for reproduction of dictation recording whose sound quality is not so important as music reproduction. Of course, in this case, it is also possible to designate and switch the reproduction of high sound quality.

For example, suppose a scene in which a recorded meeting is being played. First, reproduction is started as uncompressed information. The user can start reproduction of high-quality sound immediately after pressing a key for specifying high-quality sound, when a certain speaker makes something important or hard to hear while listening to this.

When the listening is completed, the non-compressed information may be reproduced again. As a result, current consumption can be significantly reduced. In addition, during mute reproduction, reproduction of uncompressed information may be applied since the sound quality is inconspicuous even if it is lowered.

The third effect is assuming that five pieces of music information have been recorded live for 30 minutes in a row. However, since the device side continuously records, even if five songs are entered, it is regarded as one song.

Therefore, it is necessary to specify a flag indicating an appropriate break between music pieces by some means. So-called "divide" processing is required.

This makes it possible to skip music. However, determining this divide position requires key operations at extremely delicate timing. This is because the position where the user intends to divide is slightly different from the actual position.

If there is a problem, there is a possibility that the sound part is divided. Therefore, a device equipped with a rehearsal function is provided. The rehearsal function is a function for temporarily determining the divide position and for allowing the user to confirm whether or not to divide at this position.

If the non-compressed information is reproduced according to the present invention during this rehearsal, the current consumption can be remarkably reduced.

The fourth effect is related to the third effect,
In an apparatus having a dot display, an application is conceivable in which waveforms before and after a divide are graphically displayed on a time lapse axis, the displayed waveform is confirmed, and the divide is executed.

Conventionally, to display a waveform,
Although it is necessary to decompress the compressed information, in the present invention, since the non-compressed information is recorded, it can be read and displayed.

Due to the dot resolution, information of a low sampling rate is sufficient for graphic display. Since the extension is not performed, the current consumption is reduced, and the time required for the extension is not required. Therefore, the waveform can be displayed at high speed, and the convenience is remarkably improved. Of course, a graphic display of a waveform may be performed even at times other than the divide.

The fifth effect relates to a variable-speed reproduction function of uncompressed information. According to the present invention, variable-speed reproduction can be easily provided. To perform variable speed reproduction from a recording medium on which only compressed information is recorded, it is necessary to operate the decompression circuit and adjust the time axis direction, as described in the problem to be solved by the invention.

In addition, when listening early, there is a disadvantage that the time required for decompression is too long to make it in time, so that it is unavoidable to perform thinning-out reproduction, resulting in a hard-to-hear early listening.

However, according to the present invention, uncompressed information is also recorded at the same time, and since this is used, there is no need to operate the decompression circuit. Further, the decompression circuit has a much larger processing amount per time than a general control circuit, and consumes a large amount of current because the operation clock is advanced to perform the decompression processing in real time.

[0147] This can be a fatal defect for portable devices that use batteries. Therefore, in order to reduce the current consumption as much as possible, the non-compressed information according to the present invention may be used in a scene other than the case where reproduction is actually performed with high sound quality.

The following are applied examples.

The sixth effect is that, for the reproduction of dictation recording whose sound quality is not so much as important as the reproduction of music, a method of early listening with variable speed reproduction is also conceivable. This is because the less important parts are heard quickly.

Of course, in this case, it is also possible to designate and switch high-quality sound reproduction. For example, imagine a scene playing a recorded meeting. First, reproduction is started as uncompressed information.

While listening to this, the user can start reproduction of high sound quality immediately after pressing a key for designating high sound quality if a certain speaker has made something important or difficult to hear. it can.

[0152] When the listening is completed, the reproduction may be performed in the early listening mode. As a result, current consumption can be significantly reduced.

Furthermore, there is a merit that the continuity of the sound is not impaired even in the variable speed reproduction, so that the user can easily hear the sound.

The seventh effect is that this function can be used at the time of rehearsal as described in the third effect.
On the other hand, there is a method of slow reproduction for specifying a delicate position. In addition, there is an advantage that fine positioning can be performed in the case of late listening.

This may be applied not only at the time of rehearsal, but also at the time of fast-forward playback when it is desired to find a certain portion of a song. Since the decompression circuit does not operate as described above,
There is an advantage that the processing time can be easily reduced and the current consumption can be reduced.

In addition, there is a merit that the continuity of the sound is not impaired even in the variable speed reproduction, so that the user can easily hear the sound.

The eighth effect is that, as described in the embodiment of the fast-listening playback 2 in which the pitch is not changed, when the repetition pattern of phonemes is analyzed and the fast-listening playback is performed, the analysis takes a certain amount of time. Although it is necessary to analyze with a width, according to the present invention, uncompressed information can be read over several pages without performing decompression, which is very suitable.

The ninth effect relates to the reverse reproduction function for each syllable, and the reproduction in the reverse direction can be performed while maintaining the advantages in the forward processing described in the first to eighth effects. .

Even in this case, since uncompressed information is digitized and recorded in the semiconductor memory, various processing can be performed.

A tenth effect is that although reproduction in the reverse direction produces meaningless sound as it is, reproduction in a certain unit in the forward direction allows some discrimination.

The eleventh effect is that if a means for detecting a silent part or a repetitive part of a sound is further provided, the reproduction in the forward direction is repeated while skipping in the backward direction with the part as a delimiter. In the reverse direction.

This is a process that can be easily performed because uncompressed data is recorded in advance at a low sampling rate. This makes reverse playback easier to understand.

It is also possible to record a delimiter for each syllable as a flag together with the uncompressed information at the time of recording.

The twelfth effect is not limited to the reverse reproduction as in the tenth effect and the eleventh effect, but it is of course possible to simply reproduce the data in the time series direction.
In this case, the voice cannot be heard as meaningful.

However, some discrimination is possible. At this time, if a waveform display having the fourth effect is displayed, the identification can be effectively performed.

As described above, it goes without saying that the use of the non-compressed information enables these processes to be advantageous from the viewpoint of current consumption and processing speed.

The thirteenth effect relates to simultaneous recording and reproduction, and it is assumed that, for example, a conference recording is performed. However, suppose that during the recording, it is necessary to listen to the memo recording of the telephone number that was urgently recorded before.

A situation arises in which a conference must be recorded and a memo recording must be reproduced. However, there is usually only one audio compression / decompression circuit, and it is not possible to decompress during the recording mode.

Apart from a large-scale system, especially in a battery-powered portable device that requires limited resources to be operated with limited current consumption, an audio compression / expansion circuit is used for compression and expansion. It is not very realistic to have two.

The fourteenth effect is that since uncompressed information can be reproduced even during recording, it is possible to reproduce it during recording and not only to improve convenience, but also to reduce power consumption even in a portable device. There is an effect that this convenient function can be realized without holding down the hardware and without adding hardware.

The fifteenth effect relates to the comment function, as described in the above embodiment, since it is possible to set not to simultaneously record uncompressed information during recording, so that only compressed information can be recorded first.

Alternatively, only the non-compressed information can be recorded first, and the compressed information can be recorded later. For example, a song can be recorded in the main area, and a comment such as the singer's name can be recorded as uncompressed information later.

It is also possible to record the singer's name as uncompressed information first, and then record the music in the main area.

In this way, during the cueing operation, the comment is reproduced because the non-compressed information is reproduced, and the music can be heard by switching to the compressed information, so that the convenience is improved.

A sixteenth effect is that the non-compressed information can be written in the remaining unused area where the compressed information has been written in the main area in the same page of the semiconductor memory in FIG.

Here, since it takes time to compress the audio information in a certain section, if writing is performed in accordance with the flow of information output timing, the compressed information and the non-compressed information written in the same page will take time. Uncompressed information is the information on the past side.

Of course, it is easy to write the same information temporally in the same page by the buffer processing. However, in this case, it is assumed that the recording is intentionally shifted in time without performing the buffer processing.

By doing so, unnecessary buffer processing is not required, and there is an advantage that an outline of information to be decompressed next can be known in advance during reproduction.

For example, when information to be reproduced next is almost silent, power consumption can be reduced by suspending the decompression circuit.

If the reading of the next page fails for some reason, uncompressed information is output instead, and although the sound quality changes instantaneously, continuity can be secured. There is no loss and reliability is improved.

As the seventeenth effect, the effect of the present invention has been described with a focus on the convenience of sound recording and reproduction and the current consumption. However, from the viewpoint of effective use of memory, As described in the issue on the effective use of memory, not only can the previously unused areas be used effectively, but also greater convenience and lower power consumption can be realized, greatly improving the value of the device. Can be done.

The eighteenth effect is that it is of course possible to write compressed information and non-compressed information together in one page as in the mixed writing embodiment.

For example, if it is determined that the compressed information is written alternately in the even addresses and the non-compressed information is written in the odd addresses alternately, the lowest address control signal is switched by another control signal, so that the same address management can be obtained from the host side. The non-compressed information and the compressed information can be instantaneously switched and read, so that the processing can be simplified.

In a device having a 16-bit data bus, compressed information and uncompressed information can be read simultaneously, so that data processing can be performed efficiently.

Further, according to this method, in the so-called "divide" processing, even when the page is divided at a position where the page is divided, the compressed information and the non-compressed information are similarly divided. It can be simplified.

A nineteenth effect relates to a transfer function. Since uncompressed information is managed in another area, it is possible to take out only that part and transfer it to the outside.

Thus, by extracting the uncompressed information corresponding to the enormous amount of main information, it is possible to instantaneously create a headline of a voice.

For example, a layout for editing or the like can be created based on the information transferred to a personal computer, and only necessary main information (compressed information) can be extracted. Therefore, there is an advantage that work efficiency is remarkably improved. There is.

Information can be directly transferred to a personal computer by using a removable card.

[Brief description of the drawings]

FIG. 1 is a block diagram of the present invention.

FIG. 2 is a conceptual diagram of a page according to the present invention.

FIG. 3 is a conceptual diagram of compressed information and non-compressed information according to the present invention.

[Explanation of symbols]

 K1 Key input means M0 Control circuit D1 Audio compression / expansion circuit D2 Sampling conversion circuit M1-Mn Memory chip A1 Analog input circuit A3 A / D conversion circuit

Claims (12)

[Claims] 1. A recording medium that reads and writes information in a fixed-length recording unit, digitizes and compresses and records analog data having a time axis in the recording unit, and records without compression. Recording / reproducing apparatus having means for performing 2. The recording / reproducing apparatus according to claim 1, wherein, when recording without compression, AD information acquired at a sampling rate of analog information to be compressed is thinned to a sampling rate smaller than this. A recording / reproducing apparatus comprising: 3. The recording / reproducing apparatus according to claim 1, further comprising means for simultaneously recording information obtained by compressing analog data and information decimated without performing compression. . 4. The recording / reproducing apparatus according to claim 1, further comprising means for simultaneously recording compressed information and information thinned out without compression in the same recording unit of a recording medium. Recording and reproducing device. 5. The recording / reproducing apparatus according to claim 1, further comprising means for performing reproduction based on information thinned out without performing compression. 6. The recording / reproducing apparatus according to claim 1, further comprising a variable-speed reproduction processing means such as a review and a cue for performing reproduction based on information thinned out without performing compression. Characteristic recording / reproducing device. 7. The recording / reproducing apparatus according to claim 1, further comprising a variable-speed reproduction processing unit that does not change a pitch when performing a review, a cue, or the like. 8. The recording / reproducing apparatus according to claim 1, further comprising, in a recording process at a low sampling rate, applying a variable speed reproduction process in which a pitch is not changed in advance in consideration of reproduction time. A recording / reproducing apparatus comprising means for lowering a bit rate. 9. The recording / reproducing apparatus according to claim 1, further comprising means for simultaneously processing reproduction at a low sampling rate, and compression and recording at a high sampling rate. Recording / reproducing apparatus characterized by the above-mentioned. 10. The recording / reproducing apparatus according to claim 1, wherein in the reproduction processing at a low sampling rate, the reproduction is performed in the forward direction for each phrase during the reproduction in the reverse direction. A recording / reproducing apparatus comprising means for performing reproduction in a direction going back in time. 11. The recording / reproducing apparatus according to claim 1, wherein the information thinned out without compression and the information compressed are separately managed and transferred to another apparatus. A recording / reproducing apparatus comprising means for performing. 12. An apparatus according to claim 1, wherein the recording and reading is performed by using a fixed-length recording unit write type semiconductor memory represented by a flash memory as a recording medium or a memory card. A recording / reproducing apparatus, comprising: means for performing, and means for writing compressed information and uncompressed information by effectively using a page buffer in a chip of a page write type flash memory.