JPS63253994A - Sound wave information recording/reproducing system - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a sound wave information recording and reproducing apparatus, more specifically, a method for compressing sound wave information in a recording and reproducing apparatus that records sound wave information on a sheet and reproduces the sound wave information on the sheet. Regarding.

Conventionally, records and magnetic tapes are well known as sound wave information recording media. However, these require special equipment (tape recorder) for recording and playback.

Recently, magnetic thin films have been coated on cards and sheets.
Magnetic information is written on the magnetic thin film, and letters, pictures, photographs, symbols, etc. are printed on it. In other words, one card or sheet carries both magnetic information and visible information.

Magnetic information can also be converted into audio information. However, recording both visible information and magnetic information requires two devices: a printing device and a magnetic recording device.

These two devices have completely different recording principles, and although they may be able to share the same card or sheet conveyance mechanism, they must have separate write heads and head drivers.

With the recent development of OA equipment, single sentence creation and Ig collection (
Character information processing) became widely used in word processors, and illustrations.

Inputting, modifying, creating halftone images, etc. (illustration information processing, halftone information processing) can now be easily performed using a scanner, input board or input tablet, and if necessary, additional input means such as a mouse. Advanced editing processing that combines these information processes on one screen has also been put into practical use. However, audio information is recorded on magnetic media, and is recorded using a dictation machine (
(tape recorder) required. Therefore, even if visible information and audio (sound wave) information are recorded on a single card or sheet, it is necessary to coat or paste a magnetic thin film on the card or sheet, and it is necessary to apply a magnetic thin film to the card or sheet. It requires two things: a printing press (including a copying machine) or printer, and a magnetic recording device. In view of the above-mentioned circumstances, the applicant first made it possible to visually record audio information in a reproducible manner on a card or sheet, and
We have proposed an audio information recording and reproducing device that is capable of reproducing audio information in addition to visual information such as character information, illustration information, and halftone information using one recording means.

FIG. 10 is an external view showing an example of the audio information recording and reproducing device previously proposed by the applicant, and FIG. 11 is a system diagram for explaining the operation of FIG. 10. 1
1, a word processor body 100 equipped with a CRT display 1161 and floppy disk devices 118, 119 includes microphones 1111. A speaker 1221, a keyboard 1172, an image reading scanner 1171, and a laser printer 1173 are connected. Word processor main body 100 and keyboard 1172
The combination of the laser printer 1173 and the keyboard 1173 is a so-called English-Japanese word processor.
172 to write a composition, and record the text on a floppy disk using the disk device 119 or print it out using the laser printer 1173.

The combination of the word processor main body 100, the keyboard 1172, and the scanner 1171 is a so-called image processing device, which edits the image read by the scanner as necessary, records it on a floppy disk, or prints it out.

The word processor main body 100 is connected to a local network, and the word processor main body 100, a keyboard 1172, a scanner 1171, and a laser printer 1173 constitute a so-called facsimile device.
It prints out images received from other stations (facsimile machines), and also prints out image data of originals set in a scanner, data recorded on a floppy disk, or data stored in the internal memory of the word processor main body 100. Transmit to other stations.

The combination of the scanner 1171 and the laser printer 1173 is configured as a so-called digital copying machine, but there is an image reading mode in which only the scanner operates and provides a video signal (digital) to the host (word processor body 100), and a video signal from the host. It operates in three modes: a print mode for printing out (digital) images, and a copy mode for reading original images with a scanner and printing them out.

In addition to the conventional functions described above, the above-mentioned device has functions of storing, reproducing, and printing out the sound blown into the microphone 1211, and the sound data is processed alone or in combination with image information. The functions are summarized as follows.

a6 Document creation and editing functions: You can create and edit various documents, such as using various fonts, inputting and editing images, forms, graphs, and numerical tables in addition to standard Japanese/Roman word processor functions. It has functions for creating and processing images, compositing editing them, page layout, and document formatting.

b. Audio recording editing function...The audio captured by the microphone is converted into compressed data, and this compressed data is used alone or in the above a.
This sentence is processed in combination with the basic data.

C0 audio printout function...Prints out the compressed data of the above.

d. Audio playback function: Plays back compressed data in RAM and compressed data on floppy disk to play back audio. The compressed data read by the scanner is played back into audio. In addition, the audio captured by the microphone is simultaneously played back by the speaker.

e. Function to use commercially available programs: Equipped with a function to use commercially available programs.

f. Terminal function: Equipped with a function to search for files in devices connected via the communication station, use programs, etc.

g. Print function: Equipped with a function to print created documents, audio compressed data, etc.

h. Copying function: Equipped with a function for performing normal copying.

i. Storage function: Equipped with a function to record created documents, audio compressed data, etc. on a floppy disk.

j. Search function: Equipped with a function to search for files in a file station (not shown) connected to the local network.

k.Transmission function...using the local network.

A function for sending and receiving documents, audio compressed data, etc. between other stations (not shown), and sending and receiving documents, compressed audio data, etc. between external devices via a communication station (not shown).
It has the function of transmitting and receiving compressed audio data, etc.

As shown in FIG. 11, the word processor main body 100 is connected to a local network cable 1201 via a transceiver TR at its communication control unit CCU 120.

This cable 1201 is, for example,
This is a local network cable disclosed in No. 28, to which various stations are connected. The word processor main body 100 is one workstation of this network.

Word processor body 100 which is a workstation
may be input with a mixture of audio (compressed data), documents (text, graphics), and images (pixels), or portions thereof. Here, the compressed data is data obtained by A/D converting an audio analog signal into 8-bit digital data in the digital processing circuit 121, and is
12), the data is compressed into 4-bit data through predictive encoding processing.Text is a set of coded characters, and graphics is coded graphical information.For example, 5 arcs of circles. This is a collection of commands to draw etc. Also, an image (pixel) refers to an image in pixels (
Bit string information that is divided into pixels (dot pixels) and corresponds to black and white information, brightness and color information of a pixel with bits "1" and "0", and is used by the scanner 11 in FIG.
It is input from 71. Note that from the scanner 1171,
Text and graphic information can also be entered, but at the time they are entered,
These are treated as image information. The word processor main body 100 also has functions for creating and editing documents. In other words, in addition to standard Japanese, Roman, and word processor functions, it also includes the use of various fonts, image editing, form creation, graph creation, illustration creation, creation and processing of numerical tables, composite editing of these, page layout, It has document formatting functions. In these processes, audio compressed data is also processed in the same way as these data. It also has programming capabilities and can create programs using standard languages.

It also has a terminal function and can search for files on a host computer connected via a communication station, use programs, etc.

Furthermore, it has a storage function and documents created on one main body 100 (
(including compressed audio data: the same applies below) or files,
and file station, host computer,
It has the function of storing documents or files transferred from facsimiles, other workstations, etc. on floppy disks. Note that the image information can be stored after data compression according to specifications. The main body 100 also has a search function. That is, it has a function of searching for floppy files in the main body 100, which is a file station and its own workstation.

The main body 100, which is a workstation, further has a transmission function. That is, using the resources of the network,
It has the function of transmitting and receiving documents or messages to and from other stations, such as a double image synthesizer and a facsimile machine.

Note that the image information can be compressed and transferred if specified.

As a method of inputting data to the main body 100, microphones 1211. Keyboard 1172. There are cases in which the information is input from the scanner 1171, cases in which it is input from the outside as a facsimile signal via a communication line, and cases in which it is input from the host computer.

Transfer the completed document to another workstation, send it via fax, or print it to a laser printer.
173, input processing is performed as pixel (image) information from the scanner 1171. In this case, the main body 100 handles all information in pixels (dot bits: bits in audio compressed data), compresses the data, and sends it to the network. If the recipient of the transmission is a workstation, it will be treated as an e-mail.
When the workstation is working, a message indicating that a mail has arrived is displayed on a part of the screen of the display device [1161] to notify the operator. The operator is
At an appropriate time, the document can be recalled to the screen. In this case, since the information is in pixels, the data is restored from its compressed format, subjected to linear density conversion to pixels for screen display, and then sent to the display 1161. Note that since the original compressed information remains as it is in the memory, it is determined whether to save or delete it based on the operator's instructions.

Next, as for how to use a scanner, firstly, there is the reading of various images, and secondly, there is the input of a large amount of documents.If all the images or documents are audio compressed data (usually audio from recording paper) There are three modes: playback), a part of which is compressed audio data, and a case where all of the data does not include compressed audio data.

In the word processor main unit 100, when creating a sentence that includes a mixture of voice, text, graphs (illustrations), and pictures, the voice is output from the microphone 1211, the text is input from the keyboard 1172, and the illustration is input from the keyboard 1172 (graph processing input). are respectively input from the scanner 1171 and synthesized. In that case, the main body 100 executes the image processing program to
Free editing such as moving, enlarging, and reducing the position is possible.

In order to assist this work, the image on the display 1161 is capable of multi-window processing. In other words, the image on the screen can be either a completely divided image or an image made by stacking multiple sheets of paper offset from each other. In the former, each image is reduced, while in the latter, the image is The size remains the same, but some parts are no longer visible. In addition, the system status and currently available commands are also displayed in the system area on the display.

When the compound document created in this way is stored in memory, the code (text information, illustration information) and
Stored in the form of compressed pixels (halftone images, pictures, sounds). Also, if you want to send this compound document to another workstation in the same way as the image information mentioned above,
The other workstations receive the transmitted code and compressed pixel information and decompress the code (letters) and pixels (pictures, sounds) using a character generator to display this information on the display 1161. By doing this, the pit stream is developed on the pit map memory. If this pit stream information is DMA-transferred to the video RAM 116 of the display 1161, exactly the same information as the work stage information on the sending side will be displayed on the display 116 of the receiving workstation.
1 will be displayed on top. In the same way, the file station also develops the pit stream on the bitmap memory, and the subsequent processing is the same as that for the pixel information described above to print or store the pit stream.

FIG. 12 shows the digital processing circuit 121 shown in FIG.
and a circuit diagram for explaining details of the analog processing circuit 122. The interface for A/D converting the audio signal, that is, the digital processing circuit 121 includes an amplifier that amplifies the audio signal of the microphone 1211, and an amplifier that adjusts the audio signal level. It is composed of an offset adjuster, a low-pass filter, a sample-and-hold circuit, and an A/D converter, converts the audio signal into 8-bit data, and supplies the data to the microprocessor 112 via the bus 1121.

Microprocessor 112 compresses the 8-bit data to 4 bits using predictive encoding, processes the compressed data in the same manner as pixel information, and writes it to memory.

In the case of audio playback, microprocessor 112 decodes the audio pixel information (compressed data) to obtain 8-bit data and outputs the audio signal (analog) via bus 1121.
is provided to an interface, ie, an analog processing circuit 122, for reproducing the data. The analog processing circuit 122 has 8
A D/A converter that converts bit data into an analog signal;
It is composed of a buffer amplifier, a low-pass filter, and an output amplifier, and energizes the speaker 1221 at a level corresponding to 8-bit data.

As mentioned above, the word processor main body 100 has various functions and handles a huge amount of image information and audio information, so it is equipped with a plurality of processors (CPUs). Perform parallel processing. That is, as shown in FIG.
and the sub-processor 112 are connected by a multipath 111o,
Communication between the two is performed via the main memory 113. Connection between the two is performed by an interrupt signal or a status signal.

The image data is transferred from the processor 111 to the processor 112 by
This is done via the memory 113.

Local buses 1111, 1110 are connected to both processors 111, 112, respectively, and these local buses 1111, 1110 are connected to memories 114, 115, and keyboards 1172, 1172, . Scanner 117
1 and printer 1173, encoder compressor, decoder 115
1. Image processing units 1152, 1153
, external memory or FDD (floppy disk drive unit) 1 via controllers 118 and 119
181, 1191, CRT via controller 116
A display device 1161 and a communication control device 120 are connected.

Note that the communication control device 120 connects the word processor main body 100 to the communication line, and when connected to the communication line, it is used when there is a call from the line and when the processor 111 calls the line (another station). The processor 111 is connected to the communication line.

The processor 111 (CPUI) functions as the main processor of the word processor and manages all tasks except for display tasks. Therefore, the O8 (operating system) of the word processor 100 runs on this processor 111.

Also, when the diagnostic program (Diag) is in the idle state,
nostic Program) can be run. Additionally, it has a built-in parallel I10, serial port, timer, and interrupt control circuit.

The processor 112 (CPU2) is a subordinate entity of the processor 111 (CPUI) as described above, and the CR
It operates for T-display image processing and audio data compression/playback processing. Using A/D conversion data sent from the digital processing circuit 121 (treated in the same way as pixels) and character codes, pixels, vectors, etc. sent from the processor 111 via the memory 113, the final A picture-filled document bitmap is synthesized on RAM (320KB). Then, transfer the completed picture-mixed document bitmap to the CRT controller 0-1.
16 (7) Transfer to VRAM (192KB).

When playing audio, the bitmap audio compressed data on the RAM is extracted and decoded one word (4 bits) at a time.
Bit data is created and provided to the analog processing circuit 122.

CRT controller 116 generates a water type direct synchronization signal and a video signal for a high resolution CRT display.

A VRAM is built in as a display memory, and data is transferred from the RAM in the processor 112. This controller 116 is a Japanese language landscape (Landsca).
pe) type CRT and portrait type for English text (resolution 94
5x1260 dots) monochrome raster scan system can be connected.

Memory 113 is used to transfer image data (character codes, pixels, vectors) from processor 111 to processor 112. A character generator exists in a part of the memory area. Memory address space is 1024K
It is B.

Memory 114 is the main memory of word processor 100, and has a memory address space of 1.5 MB.

It also has dual port functionality, i.e. local bus 111
Interface with 1 and parallel I10 module 1
It has an interface with 17. This transfers scanner data directly from the parallel I10 module, as well as keyboard and mouse (cursor position indication) codes.

The parallel I10117 has 12 ports (96 bits) as a parallel I10 interface, and transfers scanner data (keyboard, mouse, and scanner data) directly to the memory 114 without going through the local bus 1111.

The keyboard 1172 includes three types of character keys (for kana-kanji conversion, for tablet kanji input, and for alphabetic characters) and 16 function keys.

The scanner 1171 has a maximum reading size of A3, a resolution of 12 dots/mm (3000 PI), and can read sheet-type originals.

FDCI 1.8, 119 controls FDD (floppy disk drive unit) 1181.1191.

Floppy disks are double-sided double density (IMB/DRIV)
E) is used. In addition to storing programs, local files, kana-kanji conversion dictionaries, and character generators, these are also used as logout memories.

Once the floppy data has been read into the memory, another floppy disk for recording texts, etc. is attached to one of the disks, and written texts, etc., are recorded and recorded texts, etc. are read out.

The image processing unit (IPUI) 1151 has a binary DCR (data compression and reproduction) function.

The image processing unit (IPU2) 1152 has a function of density conversion and enlargement/reduction. The density conversion is 12 → 4 dots/mm.

12 → 6 dots/mm, 12 → 8 dots/mm, 8-+
There are 12 dots/n+m, and the enlargement/reduction is 0.
It can be set between 5 and 2 times in 0.1 steps.

Image processing unit (IPU3) 1153 has an image rotation function. Rotates +90° in one step.

The communication control device 120 controls the exchange of data transmitted via the local network, and performs protocol control including layers up to at least the data link level. In other words, data link layer control includes data capsule disassembly/assembly (framing, addressing, error detection) and link management (channel allocation (collision avoidance), collision processing), and physical layer control includes: Data encoding (preamble generation/
removal (for synchronization), bit encoding/decoding)
and channel access (C-bit transmission/reception, carrier sense, collision detection).

The transceiver TR is directly connected to the communication medium (coaxial cable) of the local network and acts as a communication control device! 1f
120 and transceiver TR are connected by a transceiver cable.

According to the above device, audio data is visibly recorded on paper or an equivalent sheet in the same way as ordinary characters or images are recorded. Although the recorded image of audio data is visible,
Normal people cannot understand the meaning. However, a widely used scanner, such as a scanner used in facsimiles, reads the image on the recording sheet to obtain audio data, and this data is reproduced into an audio signal using the reverse logic of the digital conversion described above. When the signal energizes the speaker, the recorded audio is played back. Further, according to the above device, the level of the audio analog signal is A/D converted into digital data having a relatively large number of bits, for example, 8 bits or more, and furthermore, this digital data is predictively encoded to be, for example, 4 bits or less. It uses digital conversion means to compress digital data with a relatively small number of bits, and according to this,
Digital processing of audio signals is performed using relatively simple hardware and relatively simple program processing.

Moreover, since the number of recording dots (bits) is reduced, it is possible to record audio over a relatively long period of time on a single piece of paper or a similar sheet or card (hereinafter, paper and similar sheets or cards are collectively referred to as a sheet). obtain. For example, a word processor is often used for shorthand or dictation recording, but if the manual input of the word processor is delayed during dictation, you can switch from text input to voice recording by instructing voice recording using the keyboard, or switch from text input to voice recording. and audio recording can be performed in parallel.

In addition, there are cases where it is preferable to record the author's (or dictator's) real voice instead of or together with the signature on the word processor's text, but in such cases, the word processor's CPU, etc. If the digital conversion means hardware is connected to the CPU, etc., and an audio signal/digital signal converter is incorporated into the word processor, both text and audio data can be combined and recorded on one sheet. It is also possible to record text and audio data for each sheet. The recording device is essentially a word processor and does not require an audio recording device such as a tape recorder, and since the same material sheet is used for text and audio data recording, no special audio data recording medium is required. .

Furthermore, in CAD (computer aided design), it is preferable to add the designer's explanatory voice or identification voice in addition to recorded drawings or illustrations, and even in facsimile, it is preferable to add the explanatory voice or identification voice of the designer in addition to the recorded drawings or illustrations. Instead, it is desirable that the sender's voice be automatically reproduced from the recorded data on the manuscript at the same time as the original document receiver (1) or by reproducing the audio signal from the received paper.Also,
graphs entered or read with a keyboard, input tablet, mouse, and/or image scanner;
In addition to recording illustrations, photographs, or pictures, it is preferable to record a combination of explanatory audio, background music, natural sounds, reading audio, etc., but according to the above device, these can be used for any of these. can do.

However, in the conventional apparatus described above, only the waveform is encoded, and therefore a high compression ratio cannot be obtained.

1-The present invention was made in view of the above-mentioned circumstances,
In particular, in a conventional compression method (such as DPCM) that encodes a waveform, this was done with the aim of obtaining a higher compression ratio by separating silent parts and further separating similar waveforms.

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention determines whether there is a sound or not, and if the absolute value of the input amplitude is smaller than a predetermined value in a predetermined number or more consecutively, it is regarded as a silent part, and the input amplitude is determined as a silent part. If the absolute value of the amplitude is greater than a predetermined value in a predetermined number or more consecutively, it is determined as a sound part, or in detecting the pitch of an audio waveform.

Alternately detecting maxima and minima with large absolute values of amplitude, and detecting as a pitch a deviation in a predetermined area in which the time between the maxima (or minima) is repeated a predetermined number of times or more. , or when detecting a waveform similar part after detecting the pitch of the audio waveform, if the number of extreme values in one pitch is equal to the number of extreme values in the previous pitch, the waveform in the pitch of interest is is characterized in that the waveform is similar to the waveform of the previous pitch. Hereinafter, the present invention will be explained based on examples.

FIG. 1 is a block diagram for explaining one embodiment of the present invention, in which 1 is a voiceless sound determination section, 2 is a silent time coding section, 3 is a smoothing section, 4 is a pitch detection section, and 5 is a waveform similar part detection section, and 6 is an amplitude and pitch coding section (however, in the experiments described later, the same waveform as before was used. Also, when 5 or more pieces are consecutive, waveform coating is performed for every 5 pieces) , 7 is a waveform coding section (but not compressed).

2 to 5 are flowcharts for explaining the operation of each block shown in FIG. 1, and FIG. 2 is a flowchart of the silence/sound determination unit 1, which separates silent parts and sound parts. Flowcharts, FIGS. 3 and 4 are flowcharts of the pitch detection unit 4, FIG. 3 is a flowchart of extreme value detection and pitch candidate detection, FIG. 4 is a flowchart of pitch detection from pitch candidates, and FIG. 5 is a flowchart of pitch detection from pitch candidates. , which is a flowchart of a waveform similar portion detection unit, which detects a similar portion from similar portion candidates and extreme values.

Each block in FIG. 1 will be explained below.

It is generally well known that there are always silent parts in audio data, and that similar waveforms are repeated in vowel parts6.The present invention is based on this recognition, and is based on the conventional waveform In order to obtain an even higher compression rate in a compression method that encodes a signal, silent parts are separated and similar waveforms are further separated.

Figure 2 is a flowchart of the sound/silence determination unit 1. flag.

AMP is......

TH-AMP-1 is the threshold of amplitude from sound to silence.

TH-AMP-2 is the threshold for amplitude from silence to sound.

CNTl is in the sound state and TH
- The number of consecutive amplitudes smaller than AMP-1.

CNT2 is...old...silent, TH-A
The number of consecutive pieces with a swing width larger than MP-2.

T H-CN T-1 is the threshold value of CNT-1 that changes from a sound state to a silent state.

TH-CNT-2 is...the threshold value of CNT-2 that changes from a silent state to a sound state, and the silent part has an amplitude whose absolute value is smaller than a certain predetermined value (TH-AMP-1). Predetermined number of times (TH-CN T-
1) When input continuously, it is treated as a silent part. Further, a sound part is defined as a sound part when a predetermined number of consecutive inputs (TH-CNT-2) having an amplitude whose absolute value is larger than a certain predetermined value (TH-AMP-2) is made.

As preprocessing for royal pitch detection and waveform similar portion detection, smoothing is performed to remove unnecessary high frequency components. Specifically, for example, LPF(I)=(AMP(I-1,)+2XAMP(I
)+AMP(I+1))/4. However, as a result of smoothing, LPF has A
MP is the amplitude to be input, and ■ is a time index.

Pitch Detection FIG. 3 (FIGS. 3(a) and 3(b)) is a flowchart of extreme value detection and pitch candidate detection, and FIG. 4 is a flowchart of pitch detection from pitch candidates.

In Figure 3, AMP is...
...Swinging width.

Dl is・・・・・・・・・・・・・・・・・・
...Inclination from the previous point to the point of interest.

D2 is・・・・・・・・・・・・・・・・・・
・・・・・・Tilt one position back from the point of interest.

M I N-P I TCH is the minimum possible pitch time.

In Fig. 4, P is the time (index) of the pitch candidate, and MAX-ZURE is the time between the pitch candidates of interest. The threshold value CNTP for determining whether the time and the time between the previous pitch candidate are considered to be the same pitch is......The number of consecutive pitch candidates that are considered to be the same pitch T H-CN T - PITCH is the CNTP that is the pitch.
Detect each other at the threshold of What was returned was detected as pitch.

Figure 5 is a flowchart for detecting similar parts from similar part candidates and extreme values.In Figure 1, S is...・The time (index) of the similar candidate D is the number of extreme values 5AVE
-D is the number of the previous extreme value, and when the number of extreme values in the pitch is equal to the number of extreme values in the previous pitch, the waveform in the pitch of interest is the number of extreme values in the previous pitch. It is assumed that the waveform is similar to the waveform of .

Figures 6 to 9 are diagrams showing the states of the waveforms processed as described above. Figure 6 is the original waveform, and Figure 7 is the diagram with the detected pitch. The period of time is the pitch. Figure 8 is a diagram where similar parts are detected, and the part where there is a large swing is the correct one.

This is a similar part of the waveform. Moreover, in FIG. 9, similar parts are replaced with similar waveforms up to a maximum of five consecutive times, and a-j of each waveform corresponds to each other throughout the entire series.

As is clear from the above description, according to the present invention, a high compression ratio can be obtained in audio data compression.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram for explaining one embodiment of the present invention.
FIG. 2 is a flowchart for explaining the operation of the utterance/absence determination unit, FIG. 3 is a flowchart for explaining extreme value detection and pitch candidate detection, and FIG. 4 is an operation for detecting pitches from pitch candidates. A flowchart for explaining, FIG. 5, a flowchart for explaining the operation of detecting a similar portion from similar portion candidates and extreme values, and FIGS. 6 to 9
The figure is a diagram for explaining the state of the waveform processing process.
0 is an external view showing an example of the audio information recording and reproducing device previously proposed by the present applicant, FIG. 11 is a system diagram for explaining the operation of FIG. 10, and FIG. FIG. 3 is a detailed circuit diagram of the digital processing circuit and analog processing circuit shown in the figure. 1... Silent sound determination section, 2... Silent time coding section. 3... Smoothing section, 4... Pitch detection section, 5... Waveform similar portion detection section, 6... Amplitude, pick coding section,
7... Waveform coding section. Figure 1 Figure 3 (0) ■ ■ Figure 3 (b) ■ Figure 4 ■ Figure 5 Procedural Amendment Statement 1988 Patent Application No. 88315 2, Title of Invention Sonic Information Recording and Reproducing System 3, Relationship with the case of the person making the amendment Patent applicant Ota Ri Nakamagome Address 1-3-6 Nakamagome, Ota-ku, Tokyo Name (674) Ricoh Co., Ltd. Representative Hama 1)
Hiro 4, Agent address: 1-2-7 Furocho, Naka-ku, Yokohama 231
Shear Train Yokohama No. 807 6, Drawing subject to amendment 7, Contents of amendment

Claims (9)

[Claims] (1) In the silence/sound discrimination, if the absolute value of the input amplitude is smaller than a predetermined value for a predetermined number or more consecutively, it is considered a silent part, and if the absolute value of the input amplitude is larger than the predetermined value. A sound wave information recording and reproducing method characterized in that a sound part is determined if a predetermined number or more are consecutive. (2) The sound wave information according to claim (1) is characterized in that the actual sound part is determined to be a certain predetermined time before the time when it is determined to be a sound part. Recording and playback method. (3) The sound wave information recording and reproducing method according to claim (1), wherein only the time of the silent portion is encoded. (4) In detecting the pitch of an audio waveform, local maxima and local minima with large absolute values of amplitude are detected alternately, and the time between the maximums (or minimums) is a deviation within a certain predetermined value for a predetermined number of times or more. A sound wave information recording and reproducing method that is characterized by detecting a continuously manipulated sound as a pitch. (5) According to claim (4), the value to be compared with the absolute value of the amplitude is a weighted value of the amplitude determined to be the immediately preceding large maximum (or minimum). A method for recording and reproducing sound wave information. (6) The acoustic wave information recording and reproducing method according to claim (4) or (5), wherein the input signal is smoothed as pre-processing. (7) After detecting the pitch of the audio waveform, when detecting the similar part of the waveform, if the number of extreme values in one pitch is equal to the number of extreme values in the previous pitch, the number of extreme values in the pitch of interest is A sound wave information recording and reproducing method characterized in that the waveform is similar to the waveform of the previous pitch. (8) The sound wave information recording and reproducing method according to claim (7), wherein, regarding waveform similar portions, only the number of similar waveforms is encoded, and the other waveforms are encoded. (9) The acoustic wave information recording and reproducing method according to claim (7) or (8), wherein the input signal is smoothed as pre-processing.