JPH10118111A - Voice recognizing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable the comprehension of words in voice through stimuli on the skin and to enable the comprehension of voices of conversation/television/ radio even for a deaf-and-dumb person by outputting the voices to the skin with the arrangement of plural points for pressing/vibrating/stimulating the skin. SOLUTION: This device is composed of a microphone 1 for inputting a sound, amplifier 2 for amplifying the input audio signal of the microphone 1, A/D converter 3a for converting the amplified audio signal into a digital sound, sound characteristic detecting part 3b for detecting the characteristics of the digital sound from the A/D converter 3a, word collating part 3d for determining a relevant word from the inputted characteristics by comparing and collating the characteristics of the sound from the sound characteristic detecting part 3b with the sound characteristics of words stored in a ROM 3c, signal converting part 4c for converting the word signal outputted by the word collating part 3d to the signal of an exciting part 4b, and exciting part 4b for vibrating a pressure pointers 4a according to the signal of the signal converting part.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for transmitting voice of a person's conversation, television and radio as an array of points of pressure, vibration and stimulation on the skin, and is a technique particularly useful for deaf people. .

[0002]

2. Description of the Related Art Conventional methods of communicating intentions to a deaf person include a sign language, a method of expressing a language in braille or characters by using characters and projections, and a method of contacting skin with hands and fingers. . Since the sign language / skin contact transmission method requires the intervention of a person who can perform this, a person having these abilities is required, and the voice cannot be recognized alone.
Also, the method of displaying on a CRT, liquid crystal, paper, or board in the form of characters or braille requires sophisticated equipment and human intervention, and cannot recognize voice alone.

[0003]

SUMMARY OF THE INVENTION The problem to be solved by the present invention is to solve these conventional problems, to be able to recognize speech by a deaf person alone, to have a direct conversation with a person, and to have the sound of television and radio. So that you can understand
It is an object of the present invention to provide a speech recognition device capable of bringing the social life of a deaf person closer to an ordinary person.

[0004]

Means for Solving the Problems The configuration of the present invention which has solved the above problems is as follows: 1) a microphone for inputting sound, an amplifier for amplifying the sound signal of the microphone, and an amplifier for amplifying the sound signal. A voice recognition unit that inputs the generated sound signal and determines the corresponding voice from the sound waveform, and a plurality of points that apply pressure, stimulation, or vibration to the words of the voice recognized by the voice recognition unit. 2) A voice recognition device consisting of a skin output unit that uniquely outputs the skin in an array of 2) An array of a plurality of points that apply pressure, stimulation, or vibration to the skin of a voice word is an array of Braille projection points The speech recognition device according to the above 1), which is the same as the above.

[0005]

According to the present invention, the sound collected by the microphone is amplified by the amplifier and input to the voice recognition unit, and the words of the voice included in the sound signal are determined. Output to The skin output unit expresses the word signal as a unique array of a plurality of points for applying pressure, vibration, or stimulation.
The deaf person recognizes the word from the array of points on the skin that can be pressed, vibrated or stimulated. Thereby, the words of the voice composition are transmitted by the skin sensation, so that the meaning of the voice can be understood. In particular, if the arrangement of the dots and the arrangement of the points of the Braille projections are the same, transmission can be performed in the conventional Braille format, and the practicality is enhanced.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS It is practical that a speech recognition unit for judging speech words from a sound signal converts the sound signal into a digital signal, grasps the characteristics of the sound, and searches for and determines a corresponding word. It is preferable to use a computer (CPU, RAM, ROM) or the like. It is preferable that the microphone is attached to the arm, head, and chest so that the direction in which the sound source collects sound can be changed by moving the body. As a result, the input of the voice of the other person to the specific talker can be suppressed, and there is no interference of the voice, and the understanding can be facilitated.

[0007]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is an explanatory diagram illustrating a use state of the embodiment.
FIG. 2 is a circuit block diagram of the embodiment. FIG. 3 is an explanatory diagram illustrating a skin output unit according to the embodiment. In the present embodiment, the voice recognition unit uses a computer to extract characteristics from a ROM storing the characteristics of the words of the voice, converts the amplified sound signal by a microphone into a digital signal, and converts the sound signal into a digital signal. , The characteristics of the sound are extracted, and the words included in the sound are specified by comparing with the characteristics of the ROM. The specified words are stimulated by the pressurizing needle which vibrates in the same arrangement as the Braille expression in the skin output unit, and can be recognized in the Braille expression. In the figure, 1 is a microphone, 2 is an amplifier, 3 is a voice recognition unit, 3
Reference numeral a denotes an AD converter for converting a sound signal into a digital signal, and reference numeral 3b denotes a sound characteristic detecting unit for extracting sound characteristics from the digital signal converted by the AD converter, which is realized by operation of software of a computer. Reference numeral 3c denotes a ROM storing the sound characteristics of each word of the voice "Aoi ...", and 3d denotes a sound characteristic input by comparing the sound characteristics input by the sound characteristic detection unit with the word characteristics stored by the ROM 3c. The word matching unit detects a sound word and outputs a character signal of the word. Reference numeral 4 denotes a skin output unit, 4a denotes a pressure needle having the same arrangement as the Braille dot arrangement, and 4b denotes a vibrating unit for applying pressure and light vibration of the pressure needle, similar to a printer print head of a dot printer. Having a structure. Reference numeral 4c denotes a signal converter for converting a character signal of a word output from the word matching unit 3d into a signal in a Braille arrangement of the vibration unit. Reference numeral 5 denotes a microphone, an amplifier 2, a storage casing of the voice recognition unit 3, and reference numeral 6 denotes a skin output unit 4.
Arm mounting band. In this embodiment, the microphone 1 receives the sound, is amplified by the amplifier 2, is input to the AD conversion unit 3 a of the voice recognition unit 3, and is converted into a digital signal. The characteristics for word recognition determination of the sound input by the sound characteristic detection unit 3b are detected by software of a computer, and are input to the word matching unit 3d. In the word matching unit 3d, the input characteristics are compared and compared with the characteristics of each word stored in the ROM 3c by the computer software of Cpu, and the word having the closest characteristics is determined to be a corresponding word, and the word is output as a character signal. The character signal of the word output from the word matching unit 3d is input to the signal converter 4c of the skin output unit 4, converted into a signal of the vibration unit 4b, and output to the vibration unit 4b. The vibrating unit 4b applies pressure and light vibration only to the pressing needles 4a in the Braille arrangement corresponding to the word to stimulate the skin, and transmits sound in the form of a word in the Braille arrangement with skin feeling. The position of the skin output unit 4 is not limited to the arm, but may be near the chest and the ear.

[0008]

As described above, according to the present invention, each word of the voice is transmitted in an array of a plurality of points at which the skin is pressed, vibrated, and stimulated. Can be transmitted by skin irritation. This enables deaf people to talk with others by themselves, and to understand television and radio sounds alone without translating, so that they can live a life closer to ordinary people.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram illustrating a use state of an embodiment.

FIG. 2 is a circuit block diagram of an embodiment.

FIG. 3 is an explanatory diagram showing a skin output of the embodiment.

[Explanation of symbols]

 Reference Signs List 1 microphone 2 amplifier 3 voice recognition unit 3a AD converter 3b sound characteristic detection unit 3c ROM 3d word collation unit 4 skin output unit 4a pressurizing needle 4b vibration unit 4c signal converter 5 casing 6 arm mounting band

Claims (2)

[Claims] 1. A microphone for inputting sound, an amplifier for amplifying a sound signal of the microphone, and a sound signal amplified by the amplifier for inputting a sound signal to determine a corresponding sound from a waveform of the sound. Speech recognition comprising a speech recognition unit and a skin output unit that uniquely outputs the skin in an array of a plurality of points at which the words of the speech recognized by the speech recognition unit are pressed, stimulated, or vibrated on the skin. apparatus. 2. The speech recognition apparatus according to claim 1, wherein the arrangement of a plurality of points that apply pressure, stimulation, or vibration to the skin of the word of the speech is the same as the arrangement of projection points in Braille.