JPS62167699A - Semiconductor information storage device - Google Patents

Abstract

PURPOSE:To improve the recording and reproduction of video and sound by editing the content of the 1st nonvolatile memory, writing the result to a volatile memory, reading the content of the volatile memory, confirming it and writing the content to the 2nd nonvolatile memory. CONSTITUTION:After the end of recording, all of sounds are read from a large- capacity EEPROM 1, listened by monitoring or only the reserved sound is transferred from the large-capacity EEPROM 1 to a large-capacity RAM 3 depending on the time and address. The sound rearranged, edited and stored is reproduced and confirmed out of the large-capacity RAM 3, a manual switch 5 is operated, a read command signal is sent from a write/read control circuit 6 to the large- capacity RAM 3, an address is changed from the address control circuit 4, the band is expanded by a band expansion circuit 8 via a selector 7, given to a D-A converter 9, where the signal is D-A converted and amplified by an amplifier 10. Thus, required information is written in the nonvolatile memory in an optical order. further, with respect to the video signal, it is processed similarly by using a large-capacity memory.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a semiconductor information storage device for recording and reproducing video and audio signals.

2. Description of the Related Art In conventional video and audio recording and reproducing apparatuses, video was recorded on a video tape recorder (VTR) and audio was recorded on a tape recorder.

For business use, information was recorded on records or optical discs.

Problems to be Solved by the Invention However, when recording on a VTR or tape recorder, there is a problem that the magnetic tape expands and contracts when stored for a long period of time, distorting the recorded signal, and when editing, the recorded contents of the original tape must be randomized. There was a problem that it took a long time to edit because it could not be accessed. Furthermore, records and optical discs are unsuitable for recording information in general households.

Means to Solve the Problems In the present invention, a write-once nonvolatile memory such as a large-capacity fuse ROM or one-time ROM is used as a storage recording medium, and a rewritable memory such as a large-capacity EICFROM is used. Information recorded in a non-volatile memory is read and edited into a volatile memory such as a RAM, and then once written into a write-in type non-volatile memory.

With this configuration, the information written in the large-capacity rewritable non-volatile memory can be accessed randomly, and it can be read out to the volatile memory and edited by arranging it in any order. The information can be read from the volatile memory and checked, and if it is correct, the contents can be transferred once to a write-in type non-volatile memory, written and stored, and the contents can be permanently saved.

Embodiment FIG. 1 shows the main parts of an embodiment of the present invention. First, an overview of the operation will be described. The input from the microphone 110 is A-D converted by the Mu-D converter 13, band-compressed, and stored in a large-capacity KEPR (M1), which is a rewritable large-capacity nonvolatile memory.

When reproducing the stored contents of the large-capacity KKFROM 1, the output of the large-capacity iEKFROM 1 is band-expanded by the band expansion circuit 8 and returned to its original state, then D-to-A converted by the D-M conversion 9, and then the amplifier 10
amplify it and play it through speakers or headphones. Further, the contents of the large-capacity EIFROM 1 can be read out randomly according to the address specification of the address control circuit 4 and outputted from the amplifier 10.

On the other hand, a large photo! Editing is performed by randomly reading out the stored contents of the fiIcEPROM 1 in a desired order and sequentially storing them in the RAM 3, which is a large-capacity volatile memory, from the beginning. The contents of the large-capacity RAM 3 are read out in the same manner as the contents of the large-capacity EICPROM 1 via the band expansion circuit 8 to the amplifier 10, and reproduced to confirm the editing state. After checking and if there are no errors in the edited contents, the large-capacity R
Read the contents of AM3 and transfer it to Fuse ROM2, which is a large capacity write-once nonvolatile memory, Fuse RO
The fuses of the necessary cells in the memory cells of M2 are burned out to store data. At this time, by burning out the fuse, r
Store it in IJ'. However, there is also a method that stores "0". The parts other than "1" are roJ from the beginning. As a result, the digitized audio information is written and stored in the fuse ROM 2, and since the life of the fuse ROM 2 is semi-permanent, it can be stored permanently.

Writing to and reading from each of the large-capacity IKFROM 1, large-capacity RAM 2, and fuse ROM 3 is performed by operating the manual control unit 6 to send necessary signals from the address control circuit 4 and write/continue control circuit 6 to 1.2. 3 for control.

The operation will be described in more detail below. Here, KKFROM
Let l be a removable pack format [:FROM.

When recording audio signals, for example, the frequency characteristics may be set to 15
To guarantee down to KHz it is necessary to sample the signal at least 30 x H2.

If the signal is compressed using the delta PCM method and 11 pits are used per sampling, 3X10'X
11 bits is the required capacity per second. That is, 33
A 0-bit EEFROM is required. The 1i:KPROM currently on the market has a larger capacity of 64 bits per chip, but memory is progressing rapidly and is expected to increase by 8 bits, quadrupling every three years. 1
It is expected that after o years, the size will be 43=64 times larger, that is, 4 Mbits. Therefore, if you use this, you can record approximately 12.4 seconds. Furthermore, if six chips are used together, it is possible to record for one minute.

However, RAMs with a capacity of 1M bits are already being put into practical use, and it is expected that RAMs with a capacity of 64 to 266M bits will be realized in 10 years. Therefore, if a battery-backed non-volatile RAM is used instead of the ZΣFROM chip as the large-capacity EEFROM 1, one chip can record for 776 seconds, or about 13 minutes, and four chips can record for about 52 minutes, which can be recorded in stereo. Considering this, the recording time is 26 minutes, which is sufficient for practical use. Therefore, in the following explanation, KI! :FROM
1 is a non-volatile large capacity memory, temporarily called KKFROM.
shall be.

The audio signal from the microphone 11 or another sound source is amplified by the amplifier 12, and the sample rate is 30K by the A-D converter 13).
Performs 16-bit λ-reconversion of z and converts the band compressor 14
Compress the bandwidth to 11 bits.

If the signal is a stereo signal, another system from the microphone 11 to the band compressor 14 is required. The output of the band compressor 14 is transferred to the large-capacity EEFROM 1 in 8-bit parallel or serial format. Here, it is assumed that EEFROM1 is of a serial input-serial output type. Address control circuit 4 by operating manual switch group 5
Large photo iE1! : Give a write address to FROM1. The address is 30K at the same time as the recording operation starts)
Increment by +2. On the other hand, due to the output of the write/read control circuit 6, El! : Designates FROMl to write state. Assuming that the recording time for a 1 RAN pack (4 chips included, 1 Gbit) of 'E' EPROM 1 is completed in 26 minutes, another pack is installed and further recording is performed.

After recording is complete, you can either read out all the audio from the large-capacity REPROM 1, monitor it and listen to it, or save only the parts that need to be saved due to time and address constraints to the large-capacity El! :FR
Transfer from OM1 to large capacity RAM3. Large capacity KKFR
Even if there are more than 2 packs of memory as OM1, if the portion to be saved is less than % of the total on average, use Daisha iiRAM
3, the same capacity as Daisha i EEPROM 1 is sufficient. When rearranging and editing audio signals to be stored in the large-capacity RAM 3, use the large-capacity E'E P ROM.
The memory order may be within 1 or may be random. Daisha iR
To play back and confirm the edited, rearranged and stored audio in the AM3, operate the manual switch group 5 to send a read command signal from the write/read control circuit 6 to the large capacity RAM 3, and from the address control circuit 4 to the address D while changing the band and expanding the band by the band expansion circuit 8 via the selector 7.
-A converter 9 performs D-person conversion and reads it out, and amplifier 10 amplifies it. The contents of the large capacity RAM 3 are checked and if there is no problem, the data is transferred from the large capacity RAM 3 to the fuse ROM 2. The fuse ROM 2 is a normal fuse ROM or a one-time write type ROM. Note that the reproduction control circuit 16 is
This is a regeneration control switch that determines which output from the large-capacity ICEFROM 1 to large-capacity RAM 3 is transmitted to the band expansion circuit 8.

Considering the clock speed of this circuit, for stereo, 11 x 2 = 22 bits and 22 x 30 KHz =
ae o KHz, and the contents of the large-capacity EEFROM1 to fuse ROM2 can be read out at 680K1-12. This speed is sufficient for normal large-capacity memory.

With the configuration as described above, necessary information can be written in the fuse ROM 2, which is a nonvolatile memory, in any order.

Video signals can be handled in the same way if the memory is made larger.

Effects of the Invention According to the present invention, the following effects can be obtained.

(1) Only necessary information can be easily written and stored in nonvolatile memory in a short time.

(2) Even if the transfer speed between memories is increased, the signal is handled digitally using a clock, so there is no deterioration in signal quality as in high-speed dubbing of tape.

(3) It is possible to accurately match the joints of video signals and audio signals during editing.

(4) In the case of video, the screen can be moved in the opposite direction by reversing the readout address of the signal from the memory, so forward and reverse movements can be mixed and written and stored in non-volatile memory. can be easily done.

[Brief explanation of drawings]

The figure is a specific circuit diagram of a semiconductor information storage device according to an embodiment of the present invention. 1...Large capacity rewritable non-volatile memory, 2
・・・・・・Large capacity write-once non-volatile memory, 3
...Large capacity volatile memory, 4...Address system subcircuit, 5...Manual switch group, 6...
...Write/read control circuit, 7...Selector, 8...Band extension circuit, 9...D-
A converter, 1o...Amplifier, 11...
Microphone, 12...Amplifier, 13...MuD converter, 14...Band compressor, 15...
...Reproduction control circuit.

Claims (1)

[Claims] a rewritable, large-capacity first nonvolatile memory that stores video signals and/or audio signals; a write-once second nonvolatile memory that has a smaller storage capacity than the first nonvolatile memory; Attach a large capacity volatile memory to the main body, and
A semiconductor device characterized in that the content of the nonvolatile memory is edited and written to the volatile memory, and the content of the volatile memory is read and confirmed, and then the content is written to the second nonvolatile memory. Information storage device.