JP2009080471A - Voice input personal computer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that it is necessary to incorporate a voice-character conversion software for every software since the voice-character conversion software and a word processor software are implemented by only one micro-processor unit (MPU) in a conventional voice input personal computer. <P>SOLUTION: A low rank computer is specifically used for a voice-character conversion processing section, and voice is converted to character in the low rank computer. A high rank computer MPU section is specifically used for a word processing function such as Hiragana (cursive form of Japanese syllabary)/Kana (Japanese syllabary)/Kanji (Chinese character) conversion. Thereby, it is not necessary to incorporate the voice input software into an application software such as computer aided design (CAD). <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a voice input of a conventional personal computer.

  Conventional voice input characterized personal computers have increased the computer processing speed by increasing the speed of each random access memory by software using software. As a result, the software was long, the number of samples per sound was small, and there were many conversion errors per sound.

Further, since a conventional voice input characterized personal computer is processed by one MPU, only word processing software is used.
JP 2005-188915 JP 2002-244840

  In the conventional personal computer described above, since the voice conversion process is performed by one MPU, it cannot be used at all other than word processing software, and there was a limit to using it for other software. This is because the combination of software and phonetic software is too long when combined with other software.

In the present invention, in order to achieve the above object, three random access memories are arranged in series to gradually increase the computer processing speed. Specifically, the first stage is about 1000 times and the second stage is about 1000 times. In the third stage, the computer processing speed is increased by about 100 times to about 1000 times. In this computer, only the speech characterizing process is defined, and the definition process part is performed only by the voice characterizing process.
Further, in order to solve the problem, existing software can be used with a conventional MPU as a host computer, and word processor software, accounting books, CAD, CG, etc. can be input by voice.

  As described above, in the present invention, after collecting the sound with a microphone, it passes through an amplifier with an auto volume control, removes an extra component of the sound with a filter, and different sound quality for each person, and digitizes it with an analog / digital converter. In the circuit for detecting the so-called loose voltage drop portion between phrases, or in the circuit for detecting where the bit becomes 0 bit, the so-called loose inter-phrase analog-to-digital converter in the detection circuit, the electronic circuit, that is, every sound In the inter-phrase discrimination circuit, it is separated by an electronic switch, for example, input to the first stage of the random access memory, switching the synchronization signal or timing pulse, outputting at a speed of about 1000 times, and further random at the second stage by the electronic switch In the access memory, the synchronization signal or timing pulse is switched by the electronic switch in the same way as above, and the speed is about 1000 times In addition, the third stage random access memory has a speed of about 100 times to about 1000 times, and the digitized voice numerical value for each sound by the inter-phrase discriminating circuit or the detecting circuit between the phrases for each sound is subordinate to the lower computer. Compared with the existing phonetic digitization model, the voice is converted into speech by recognizing each sound.

  As described above, in the phoneticized personal computer of the present invention, the phoneticization is also performed by a lower computer, that is, a multi-computer system. By using a lower computer having only phonetic characters, the upper MPU or the computer also has existing software, that is, word processor software. It can easily be used for accounting books, calculation programs such as CAD, CG, etc. using the included word processing software.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS.

  In FIG. 1, the sound is collected by the microphone 1, the sound is adjusted to a constant volume by the amplifier 2 and the auto volume control 2, and the excess component of the sound is removed from the sound by the filter 3. A detection circuit 9 for detecting an interval between phrases of each sound by the analog / digital converter 4 and detecting a portion having the smallest digital amount from the analog / digital converter 4 as an interval between phrases. It is also an inter-phrase discrimination circuit, and an electronic switch outputs it at a speed of 1000 times by an input / output speed conversion circuit and random access memory 6a as a single sound or a few plural sounds, and then the electronic switch 5b In the circuit, the instruction or signal is output from the inter-phrase detection circuit 9 for each sound, the circuit is opened, and the output is increased 1000 times in the secondary random access memory 6b. Further, the electronic switch 5c detects the interval between phrases and opens the circuit to increase the speed to the next input / output speed conversion random access memory 6c by about 100 times to about 1000 times. A signal is sent from the detection circuit 9 to the high-speed output circuit 10 or an instruction signal is sent to convert the timing pulse or synchronization signal of the memories 6a, 6b, 6c, and the signal is sent from the high-speed output circuit 10 to the random access memories 6a, 6b, 6c. That is, when a timing pulse or a synchronization signal is input and output, a timing pulse or a synchronization signal having a speed of about 1000 times is output, and a digital sound for each sound digitized is output at high speed. The voice digitized in the definition process 7 is converted into characters in synchronization with the processing speed of the computer, and the original MPU 8 To documented by pro software, and printed or displayed on the display 11 or printer 12.

  Further, the software is also considered for software to achieve the above-mentioned purpose, and the convenience is increased by subdividing the OS in FIG. 2, and further, it is a violent software, that is, a calculation program. If 17a is the communication OS and the spread 2 is only language program conversion, the violent software program is an assembler program or normal microprogramming. Specifically, hackers such as computer viruses, pennies, and winnies are composed of calculation programs. Therefore, in a language-only language conversion program, even if a computer program enters a calculation program such as a computer virus, only the language is displayed or the computer does not operate. Therefore, a calculation program such as a computer virus is displayed or displayed. Print doesn't work anymore, Computer viruses etc. consist of calculation formulas, specifically computer viruses perform multiplication etc. to infinity, and other violent software programs such as Winnies and Penny etc. also consist of computation programs, and the communication OS is a language By making a conversion program, all the violence software will be able to display computer programs and not be able to do violence on the computer, and the spread 2 OS and the single S1 to S6 will be language conversion only and dedicated to word processing software. Music, video, etc. can be processed from S1 to S6.

  As described above, S1 may be a communication OS, S2 to S6 and the like may be, for example, a word conversion software for a language conversion program, S3 may be a word processing software for a language conversion program, and spread 3S7 may be an accounting OS for only four arithmetic operations. In the violent software that consists of computer virus and other calculation programs, the operation is only in the four arithmetic operations, and the violent software operates only a little.

  Furthermore, S13 to S17 of spread 4 established by the calculation program is calculated by configuring functions, differentiation, integration formulas, etc. as a single calculation program OS as CAD, CG, chemistry, physics, structure calculation, fluid dynamics etc. as S13 to S17 etc. If the independent OS is used as a program, computer viruses, winnies, pennies, etc. that have entered communication OS, S1, etc. cannot operate. In this way, if the spread 2 is used only as a language conversion program, violent software cannot be operated. Moreover, if only the four arithmetic operations from S7 to S12 of the spread spread 3 and the language conversion program, the violent software only operates the four arithmetic operations, so the violent software uses mathematical formulas other than the infinite four arithmetic operations, They simply display language, numbers and infinite mathematical formulas as characters and cannot operate or spread , 16 S13 to S17, when combined with a language conversion program and a program containing a calculation program or a mathematical expression, the violence software operates, but if the communication program is spread and the language conversion program of only S1, 17a of 2,14, If the software does not operate and each OSS1 to S17 is independent, the safety is further increased because it is independent. Further, S18 and 26 are used as erasure programs. Here, the spreads 1 and 13 are start pages.

  Using FIG. 3, the word processor software, that is, the upper-level software having only the spread two-language conversion program is made into a certain style, and subdivided into SA1 to SA8, 21 using SA, 20 on S2, 17b, If the execution is set to 22 and the return is set to 23, and the method is styled and all the same method styles are used, it can be used in one way, so it can be used easily. 21 is divided into headings as in SA and 20, and if the same usage method is used, the same usage method and the same usage method can be learned, and all the same usage methods are obtained. To change the font or character size with, go back to the heading in the middle of the sentence, specify the font or character size or color, etc. It is possible to change the program of instruction.

  The OS composed of the spread three language conversion program of FIG. 4 and the OS of only four arithmetic operations is suitable for books of only four arithmetic operations such as accounting books, inventory management, production management, etc. The upper software is placed on top and subdivided, and the usage pattern is subdivided one by one, specifically on the upper software SG, 24 on target SG1, 24 to SG8, 25 on S7,18. 22 and return 23, depending on the case, the number of subdivisions may be increased. If subdivision software is further created in the lower part of the subdivision part and used in the same pattern, the same method, the same method, etc. can be used in one way. If you remember, look at the target menu with the heading SG, 24 and specify it from the target menu.

The block diagram of the voice input character conversion personal computer which shows embodiment of this invention OS diagram that simplifies the use of the same voice input characterized personal computer Language conversion program diagram with the upper software of the OS diagram Combination program diagram of four arithmetic operations and language conversion program of the upper software of the OS diagram

1 Microphone 2 Amplifier with auto volume control 3 Filter 4 Analog / digital converter 5a Electronic switch 5b Electronic switch 5c Electronic switch 5d Electronic switch 6a Random access memory 6b Random access memory 6c Random access memory 7 Constant processing, lower computer 8 Computer 9 Detection Circuit (inter-phrase discrimination circuit)
10 High-speed output circuit
11 Display, display device
12 Printer
13 OS spread 1
14 OS spread 2
15 OS spread 3
16 OS spread 4
17a Communication OS language conversion program only
17b Language conversion program only
18 Language conversion program only and arithmetic operation program
19 Language conversion and calculation programs
20 Headings such as word processing software
21 Subdivisions of word processing software
22 execution
23 Execution to return
24 Headings for combinations of language conversion programs and arithmetic operation programs
25 Subdivided parts of language conversion program and arithmetic operation program ・ Erasing program

Claims (1)

The sound is collected with a microphone, amplified with an amplifier, the volume of the sound is made constant with auto volume control, the excess component of the sound is removed with a filter, and the sound is digitized with an analog-digital converter. The sound is detected to a certain degree by a detection circuit, and is input to the first random access memory for each sound, for example, and the output speed is increased by about 1000 times for each sound, and the output speed is set as a synchronization signal or With the timing pulse, the sound output for each sound is increased to about 1000 times in the second random access memory, and the synchronization signal or timing pulse is switched in the same manner as described above, and about 100 times to about 1000 in the third random access memory. The signal for each sound, that is, the digitized voice signal is recorded on the definition processing computer at the processing speed of the definition processing computer. And character of the voice and down recognition, the character of the signal for each one sound at word processing software in a conventional MPU i.e. host computer, to sentences character of.

Images