JPH06139791A - Sound storing and reproducing device - Google Patents

Abstract

PURPOSE:To enable using a device recognizing a data storing region with the same sense as a tape recorder for sound by storing remaining sound data after EOD. CONSTITUTION:Sound to be stored is divided into sound data indicating a sound part and silence data indicating time length of a silence part and they are stored, and sound is reproduced based on stored data in reproducing operation. A (c) part corresponding to time length of old sound data out of new sound data to be stored is stored in a region 61 from a memory top address MTA, and out of the remaining sound data, a (b) part is stored in a region 64 after end of data EOD and a (d-c-b) part is stored in a region 63.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice storage / playback apparatus, and more particularly to a voice storage / playback apparatus for storing voice by dividing it into voiced data and silent data.

[0002]

2. Description of the Related Art Conventionally, in an audio storage / reproduction device using a semiconductor memory such as an IC card, an audio signal to be stored is converted into a digital signal and then framed in a fixed format at fixed time intervals. The framed data is stored in the semiconductor memory in frame units. At that time, a demarcation mark (hereinafter referred to as “E mark”) is input for each sentence or phrase, and a document (or “letter”) is generated.

On the other hand, in recording a silent portion of a voice, data is not stored in frame units as described above, but instead, data indicating the number of silent frames is stored in a semiconductor memory. As a result, the silent portion can be stored without consuming the storage capacity of the semiconductor memory. In the reproducing operation, the silent portion is reproduced by inserting a frame time corresponding to the number of stored silent frames between the sound portions.

FIG. 6 is a data storage mode diagram showing a data storage mode of a semiconductor memory in a conventional audio storage / reproduction device. Referring to FIG. 6, in a document,
It is assumed that old data OD for the total reproduction time a from the point X on the way to the E mark indicating the end of the document is stored. This old data OD is assumed to include c voiced data from point X to X'point, b voiceless data from point X ', and f voiced data from point X'to E mark. . Therefore, a = c + f +
b holds.

Now, consider a case where new data ND is written over old data OD that is no longer needed. In the example shown in FIG. 6, the new data ND is assumed to include d voiced data starting from point X. X new data ND
By writing from the point, the final address of the data reaches the Y point, and therefore the E mark of the old data OD becomes Y.
A data area corresponding to e from the point is left. Therefore, as the new data ND, data for (d + e) is stored from the point X to the E mark.

[0006]

Therefore, as can be seen from FIG. 6, the old data OD and the new data ND occupy the same storage area in the semiconductor memory.
Due to the presence of b-less data included in the old data OD, a> d + e, and the reproduction time may differ even though the storage data length is the same. On the contrary,
When the new data ND contains b or more silent data, a <d + e holds, and even in this case, the reproduction time may differ even though the storage data length is the same.

As described above, although the data having the same data length is stored in the semiconductor memory, the reproduction time when the data is reproduced is different, so that, for example, a conventional audio tape recorder is used. There is an inconvenience that the voice storage / playback apparatus cannot be used with such a feeling. That is, for example, recognizing that the old voice having a length of a is stored, the new voice is a
Even if only the minute data is stored, the old data O may be deleted due to the silent data existing in the old data OD or the new data ND.
New data ND may be stored beyond the storage area of D, or new data ND may be stored with the storage area of the old data OD remaining in some cases. As a result, other old data may be lost due to the new data ND, or unnecessary old data may be left behind. That is, it was not possible to use this audio storage / playback device as if using a conventional audio tape recorder.

The present invention has been made to solve the above problems, and provides a voice storage / playback apparatus capable of recognizing and using a data storage area in the same manner as a voice tape recorder. With the goal.

[0009]

SUMMARY OF THE INVENTION A voice storage / playback apparatus according to the present invention indicates voice with voiced data representing a voiced portion, a voiceless flag indicating the position of a voiceless portion, and a time length of the voiceless portion. It includes a conversion means for converting into silence data and a data storage means for storing the voiced data, the silence flag and the silence data. The data storage means stores old voice data and old silence flags at consecutive addresses by storing old voice. This audio storage / playback device further includes
In a new voice storage operation, a silence flag detection means for detecting an old silence flag in the data storage means, and a data length of new voiced data included in the voice to be newly stored in response to the silence flag detection means. Data length comparison and detection means for detecting that the length is longer than the data length of the old voiced data, and in response to the data length comparison and detection means, the data length of the old voiced data among the new voiced data is detected. First writing the corresponding data into the old voiced data storage portion in the data storage means
Of the new voiced data and the new silence data, the remaining data not written by the first writing means is the data unstored area in the data storage means for the data length of the old silence data. And second writing means for writing the data into the data unwritten area by the second writing means.

[0010]

In the voice memory reproducing apparatus according to the present invention, the first
The writing means writes the data corresponding to the data length of the old voiced data in the new voiced data to the old voiced data storage portion in the data storage means, and then the second writing means Of the sound data and the new silence data, the remaining data not written by the first writing means is written in the data unstored area in the data storage means by the data length of the old silence data. The data write information to the data unstored area by the second writing means is stored. As a result, the recording time can be recognized in the same sense as that of the audio tape recorder, and new audio can be stored.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is a block diagram of an audio storage / playback apparatus showing an embodiment of the present invention. Referring to FIG. 1, a voice storage / playback apparatus 1 includes a removable memory card 6 having a semiconductor memory (not shown) for storing voice data. In the voice storing operation, the voice to be stored is converted into an electric signal by the microphone 2, and the converted voice signal is amplified by the amplifier 3. Amplified voice signal is A
It is converted into a digital signal by the D converter 4. The voice compression circuit 5 compresses the converted voice data by dividing the voice data into a voiced portion and a voiceless portion, and the voiced data and the voiceless data are written in the memory card 6 as data DAT.

In the voice reproducing operation, the voice data DAT stored in the memory card 6 is given to the voice expanding circuit 8. The voice expansion circuit 8 expands the compressed data by combining the voiced data and the silent data. The AD converter 9 converts the expanded data into an analog signal. The reproduced analog audio signal is amplified by the amplifier 10 and then given to the speaker 11.
The reproduced sound is emitted through the speaker.

The audio storage / playback apparatus 1 further includes a system microcomputer 7 for controlling the above operation, a key switch 12, and a display 13. The system microcomputer 7 is a semiconductor memory (not shown) in the memory card 6.
The memory card 6 is provided with an address signal ADR for controlling access to the memory card 6, a write enable signal WE, a chip selection signal CE, an erase signal ERS, and the like.

The system microcomputer 7 uses the address signal AD
While incrementing R, the voice data given from the voice compression circuit 5 is repeatedly written in the memory card 6 one after another.

FIG. 2 is a block diagram of a semiconductor memory provided in the memory card 6 shown in FIG. As a semiconductor memory, for example, a batch erase type EEPROM called a flash memory is used. With reference to FIG. 2, this semiconductor memory includes a memory cell array 20 including memory cells (not shown) arranged in 1024 rows × 512 columns.
A register 21 connected to a bit line in the memory cell array 20, a row decoder 23 for selecting a word line in the memory cell array 20, and a column decoder 24 for selecting a column to be accessed in the memory cell array 20. Including. The memory cell array 20 is divided into a total of 128 blocks each including eight memory cell rows. A total of eight such memory cell arrays 20 are provided, and a storage capacity of 4 megabits can be obtained.

In the batch erasing type EEPROM, erasing can be performed for all bits or blocks, and data writing can be performed for each byte. Erase and write are performed electrically. A register 21 is provided for high-speed input / output of data for one row in page mode. Register 2
1 is composed of a 512-bit SRAM. One block 22 corresponds to an erase unit, and the block size is 512 × 8 = 4 Kbytes.

Each of FIGS. 3 to 5 is a writing mode diagram showing an example of a mode of writing data to the semiconductor memory in the audio storage / reproducing apparatus shown in FIG. Referring to these figures,
First, the manner of writing data to the semiconductor memory in the audio storage / playback apparatus will be described. 3 to 5
Also in each of the cases shown in FIG. 6, it is assumed that the previously described data OD shown in FIG. 6 is stored as the old audio data OD. That is, the old data OD is X
It consists of the first half of the voiced data for c minutes starting from the point, the b of silence data at the point X ', and the second half of the voiced data for the f portion from the point X'to the E mark, and the total a.
It is assumed that data corresponding to minutes (= c + b + f) is stored in the semiconductor memory as old audio data OD.

Case 1 (c <d <c + b) As shown in FIG. 3, of the voiced data included in the new voice data ND, data corresponding to c is stored from point X to point X '. Of the new voiced data, the remaining (dc) data is the end of data EOD as shown in FIG.
Will be remembered after. Further, after the storage of (dc), the silence data of (c + bd) is stored after the end of data EOD. After this, the index is X '
The index operation is set to return to the point.

Case 2 (c <c + b <d) In this case, as shown in FIG.
After the voiced data for the first c of D is stored from the point X to the point X ', the voiced data for the remaining b is stored after the end of data EOD. Furthermore, the remaining (d-
The voiced data for c-b) is stored after point X '.

Case 3 (when d <c and the new data ND includes a silent part) It is assumed that the new data ND has d sound data and g sound data. As shown in FIG. 5, d voiced data of the new voice data ND is stored from point X, and then g voiceless data is stored. g
After the silent data for one minute is stored, the index is (d +
The index operation is set to jump after the X ″ point after g) minutes.

FIG. 7 is a flowchart of the audio data writing operation in the audio storage / playback apparatus shown in FIG. The data writing mode shown in FIGS. 3 to 5 is executed according to the flow chart shown in FIG.

Referring to FIG. 7, in step 41,
It is determined whether the old audio data OD includes silent data. That is, b> 0 is determined. b>
When it is 0, the process proceeds to step 42. If there is no silent data (b = 0), the process proceeds to step 43, and the d-sound data included in the new data ND is stored from the point X.

In step 42, the old audio data O
The length c of the voiced portion of D is compared with the length d of the voiced portion of the new audio data ND. When c <d, the process proceeds to step 44. When c> d, the process is step 5
Go to 2.

In step 44, the new data ND
Data corresponding to c minutes of the voiced data included in is stored. Thereafter, in step 45, address information for instructing the jump to the end of data EOD is stored in the index area in the semiconductor memory.

In step 46, the time length d of the voiced data included in the new voice data ND is compared with the sum (c + b) of the time lengths of voiced data and silence data included in the old voice data. When d> c + b, the process proceeds to step 47. Conversely, when d <c + b, the process proceeds to step 50.

The processing of steps 47 to 49 corresponds to the above case 2. That is, in step 47,
As shown in FIG. 4, the remaining voiced data of the new voice data ND from the end of data EOD is stored by b. Thereafter, in step 48, address information for jumping to the point X'is stored in the index area of the semiconductor memory. Further, in step 49, the last voiced data included in the new voice data ND from point X'is stored for (d-c-b).

The processing in steps 50 and 51 corresponds to the above case 1. That is, in step 50, as shown in FIG. 3, after the end of data EOD, the remaining voiced data of the new audio data ND is (dc).
Minutes, and then (c + b-d) silent data is stored. Thereafter, in step 51, address information for instructing the jump to the point X'is stored in the index area in the semiconductor memory.

The processing of steps 52 and 53 corresponds to the above case 3. That is, in step 52, as shown in FIG. 5, after the voiced data included in the new voice data ND is stored for d minutes, the silence data is stored for g minutes. In step 53, address information for jumping the index to the X ″ point is stored in the index area in the semiconductor memory.

The processing shown in FIG. 7 is performed under the control of the system microcomputer 7 shown in FIG. 1, so that data writing is performed in the modes 1 to 3 shown in FIGS. 3 to 5. .

FIG. 8 is an address map in the semiconductor memory shown in FIG. Referring to FIG. 8, the address in the semiconductor memory is the audio data area 3 between the memory top address MTA and the memory bottom address MBA.
0a and an index area 30b. The voice data area 30a includes an area 31 in which data of the first continuous voiced portion of the stored voice is stored, an area 32 in which an index jump flag JF1 indicating the presence of a silent portion is stored, and the remaining voiced area. It includes an area 33 for storing the data of the sound part and a data non-storage area 34 which is not yet used for storing the data. Therefore, regions 33 and 3
The end of data EOD exists at the boundary with 4. If the silent portion is included in a plurality of places, a plurality of areas for storing the index jump flag JF1 are present in the audio data 30a. Figure 8
In the example shown in (1), there is only one silent part.

The index area 30b includes an area 35 in which the index data ID1 is stored. Index data ID1 is also index jump flag JF
The number of 1's is provided in the index region 30b.

If the stored voice data includes one silent portion, the data read from the semiconductor memory will be as shown in FIG.
Is performed as follows. After the playback of the stored voice is requested, the memory top address M shown in FIG.
The voiced data 31 stored in order from TA is read. After all the voiced data 31 has been read, the index jump flag JF1 is read. When the index jump flag JF1 is detected, the index data ID1 stored in the index area 30b is referred to (arrow AR1).

The index data ID1 includes information for reproducing the silent period for a time length corresponding to the number of frames in the silent portion. After this silent time is reproduced, the voiced data in the area 33 is successively read (arrow AR2).

FIG. 9 is an address map for explaining the interlaced write operation of data to an address after the end of data EOD. The address map shown in FIG. 9 is
This corresponds to the write operation in case 2 shown in FIG. However, in order to simplify the explanation, it is assumed that the point X shown in FIG. 4 corresponds to the memory top address MTA in FIG.

Referring to FIG. 9, sound data for c of new audio data ND is stored in area 61.
The area 62 stores the index jump flag JF2. When the index jump flag JF2 is detected, the index data ID2 as the interlacing operation information stored in the area 66 in the index area 30b is referred to (arrow AR11). By referring to the index data ID2, after the voiced data for b is stored in the area 64 after the end of data EOD (arrow AR12), the remaining (d-c-
The sound data for b) is stored in the area 63 (arrow AR13). In this way, the new audio data ND is written in the semiconductor memory in the manner shown in FIG.

As described above, since the new audio data ND is written in the semiconductor memory in the audio memory reproducing apparatus 1 shown in FIG. 1 in the writing mode shown in FIGS. Even if silent data is included in the OD, new voiced data exceeding the time length c minutes of the old voiced data is stored after the end of data EOD.
Therefore, since the old audio data OD and the new audio data ND can be temporally matched in writing the audio data to the semiconductor memory, the audio storage / playback apparatus can be used in the same manner as an audio tape recorder. .

[0037]

As described above, according to the present invention, of the new voice data and the new silence data, the remaining data which has not been written by the first writing means has the data length of the old silence data. Since the second writing means for writing the data in the data non-storage area in the data storage means is provided, the data storage area can be recognized and used with the same feeling as that of the audio tape recorder. A memory reproducing device is obtained.

[Brief description of drawings]

FIG. 1 is a block diagram of an audio storage / playback apparatus showing an embodiment of the present invention.

FIG. 2 is a block diagram of a semiconductor memory provided in the memory card 6 shown in FIG.

3 is a writing mode diagram showing an example of a mode of writing data to a semiconductor memory in the audio storage / playback device shown in FIG. 1. FIG.

4 is a writing mode diagram showing another example of a mode of writing data to a semiconductor memory in the audio storage / playback device shown in FIG. 1. FIG.

5 is an audio aspect diagram showing still another example of an aspect of writing data to a semiconductor memory in the audio storage / playback device shown in FIG. 1. FIG.

FIG. 6 is a writing manner diagram showing a manner of writing data to a semiconductor memory in a conventional audio storage / playback apparatus.

7 is a flowchart of an audio data writing operation in the audio storage / playback device shown in FIG. 1. FIG.

8 is an address map for explaining a silent data reproduction operation in the audio storage / reproduction device shown in FIG. 1. FIG.

9 is an address map for explaining an interlaced write operation of data to an address after end of data in the audio storage / playback device shown in FIG. 1. FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 audio storage / playback apparatus 4 AD converter 5 audio compression circuit 6 memory card 7 system microcomputer 8 audio decompression circuit 9 DA converter 30a audio data area 30b index area

Claims (1)

[Claims] 1. A voice storage / playback device for storing and playing back a voice including a voiced part and a voiceless part, the voice comprising voiced data representing the voiced part, and a voiceless flag indicating the position of the voiceless part. , A conversion unit for converting into silent data indicating the time length of a silent portion, and a data storage unit for storing voiced data, a silence flag and silence data, the data storage unit, by the storage of old voice, Old voiced data and old silence flag are stored at consecutive addresses, and in a new voice storage operation, a silence flag detection means for detecting the old silence flag in the data storage means, and a response to the silence flag detection means. And a data length comparison and detection means for detecting that the data length of the new voiced data included in the voice to be newly stored is longer than the data length of the old voiced data. A first book for writing data corresponding to the data length of the old voice data in the new voice data to the old voice data storage portion in the data storage means in response to the data length comparison and detection means. And a remaining data not written by the first writing means among the new voice data and the new silence data, the data non-storage area in the data storage means by the data length of the old silence data. An audio storage / playback device comprising: a second writing unit that writes the data into the data writing unit; and a unit that stores the data writing information by the second writing unit to the data unstored area.