KR970037890A - Car Speech Recognition Device - Google Patents

Car Speech Recognition Device Download PDF

Info

Publication number
KR970037890A
KR970037890A KR1019950055190A KR19950055190A KR970037890A KR 970037890 A KR970037890 A KR 970037890A KR 1019950055190 A KR1019950055190 A KR 1019950055190A KR 19950055190 A KR19950055190 A KR 19950055190A KR 970037890 A KR970037890 A KR 970037890A
Authority
KR
South Korea
Prior art keywords
voice
index
output
frequency
signal
Prior art date
Application number
KR1019950055190A
Other languages
Korean (ko)
Other versions
KR100262564B1 (en
Inventor
심갑종
Original Assignee
전성원
현대자동차 주식회사
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 전성원, 현대자동차 주식회사 filed Critical 전성원
Priority to KR1019950055190A priority Critical patent/KR100262564B1/en
Publication of KR970037890A publication Critical patent/KR970037890A/en
Application granted granted Critical
Publication of KR100262564B1 publication Critical patent/KR100262564B1/en

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G10L2021/02087—Noise filtering the noise being separate speech, e.g. cocktail party

Abstract

본 발명은 차량용 음성 인식 장치에 관한 것으로서, 음성을 인식하여 전기 신호로 바꾸어 출력하는 마이크와; 상기 마이크에서 출력되는 아날로그 음성 신호를 입력받아, 디지탈 신호로 변환하여 출력하는 A/D 변환기와; 상기 A/D 변환기에서 디지탈 변환된 음성 신호를 입력받아 주파수를 분석하여, 주파수 분석된 음성 데이타를 출력하는 주파수 분석수단과; 상기 주파수 분석수단에서 주파수 분석된 음성 데이타를 입력받아 유사한 패턴의 대표값으로 묶어 대표값에 인덱스를 붙이고 인덱스 시퀀스를 출력하는 벡터 퀀타이저와; 상기 벡터 퀀타이저에서 출력되는 인덱스 시퀀스를 입력받아 음성 인덱스만을 검출하여 출력하는 인덱스 비교수단과; 상기 인덱스 비교수단에서 출력되는 음성 인덱스를 입력받아 음성 명령의 인식 결과를 출력하는 인식 판단수단으로 구성되어, 차량 노이즈가 상당히 큰 상황에서 음성 명령을 추출하여 인식을 수행함으로써 차량의 주행 소음에 대해서 정확한 음성 명령을 검출하고, 음성 구간 판별을 위해 별도의 연산을 필요로 하지 않아 연산 시간을 단축시키며, 쉽게 구현 가능하고, 메모리의 용량을 줄이는 효과를 가진 차량용 음성 인식 장치에 관한 것이다.The present invention relates to a vehicle voice recognition device, comprising: a microphone for recognizing a voice and converting the voice into an electrical signal; An A / D converter which receives an analog audio signal output from the microphone, converts the digital audio signal into a digital signal and outputs the digital signal; Frequency analysis means for receiving a digitally converted voice signal from the A / D converter and analyzing a frequency to output frequency-analyzed voice data; A vector quantizer which receives the voice data analyzed by the frequency analyzing means, binds the representative data with similar patterns, indexes the representative values, and outputs an index sequence; Index comparison means for receiving an index sequence output from the vector quantizer and detecting and outputting only a voice index; Recognition determination means for receiving the voice index output from the index comparison means and outputs the recognition result of the voice command, and extracts the voice command in the situation where the vehicle noise is quite large to perform the recognition to accurately drive noise of the vehicle The present invention relates to a speech recognition device for a vehicle that detects a voice command and does not require a separate operation to determine a voice section, thereby shortening the calculation time, easily implementing the same, and reducing the memory capacity.

Description

차량용 음성 인식 장치Car Speech Recognition Device

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음Since this is an open matter, no full text was included.

제1도는 종래의 차량용 음성 인식 장치의 구성도.1 is a block diagram of a conventional vehicle speech recognition device.

제2도는 이 발명의 실시예에 따른 차량용 음성 인식 장치의 구성도.2 is a block diagram of a vehicle speech recognition device according to an embodiment of the present invention.

제3도는 이 발명의 실시예에 따른 인덱스 비교부에서의 음성 구간 검출 방법의 흐름도.3 is a flowchart of a voice interval detection method in an index comparison unit according to an embodiment of the present invention.

제4도는 이 발명의 실시예에 따른 인덱스 비교부에서의 음성 구간 검출 방법의 실행 결과를 나타내는 도면.4 is a diagram showing an execution result of a voice section detection method in an index comparison unit according to an embodiment of the present invention.

Claims (2)

음성을 인식하여 전기 신호로 바꾸어 출력하는 마이크와; 상기 마이크에서 출력되는 아날로그 음성 신호를 입력받아, 디지탈 신호로 변환하여 출력하는 A/D 변환기와; 상기 A/D 변환기에서 디지탈 변환된 음성 신호를 입력받아 주파수를 분석하여, 주파수 분석된 음성 데이타를 출력하는 주파수 분석수단과; 상기 주파수 분석수단에서 주파수 분석된 음성 데이타를 입력받아 유사한 패턴의 대표값으로 묶어 대표값에 인덱스를 붙이고 인덱스 시퀀스를 출력하는 벡터 퀀타이저와; 상기 벡터 퀀타이저에서 출력되는 인덱스 시퀀스를 입력받아 음성 인덱스만을 검출하여 출력하는 인덱스 비교수단과; 상기 인덱스 비교수단에서 출력되는 음성 인덱스를 입력받아 음성 명령의 인식 결과를 출력하는 인식 판단수단으로 이루어지는 것을 특징으로 하는 차량용 음성 인식 장치.A microphone for recognizing voice and converting the voice into an electrical signal and outputting the voice signal; An A / D converter which receives an analog audio signal output from the microphone, converts the digital audio signal into a digital signal and outputs the digital signal; Frequency analysis means for receiving a digitally converted voice signal from the A / D converter and analyzing a frequency to output frequency-analyzed voice data; A vector quantizer which receives the voice data analyzed by the frequency analyzing means, binds the representative data with similar patterns, indexes the representative values, and outputs an index sequence; Index comparison means for receiving an index sequence output from the vector quantizer and detecting and outputting only a voice index; And a recognition determining means for receiving a voice index output from the index comparing means and outputting a recognition result of the voice command. 제1항에 있어서, 상기한 인덱스 비교수단은 상기 벡터 퀀타이저에서 출력되는 인덱스 시퀀스를 입력받아 판독한 후에, 입력된 인덱스가 잡음 인덱스인지를 판정하고, 잡음 인덱스가 아니라면, 신호의 크기를 경계치와 비교하여 음성 인덱스인지를 판단하여 음성 인덱스만을 검출하여 출력하는 것을 특징으로 하는 차량용 음성 인식 장치.The method of claim 1, wherein the index comparison means receives and reads an index sequence output from the vector quantizer, and then determines whether the input index is a noise index. And a voice index is detected and outputted by determining whether the voice index is a voice index. ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019950055190A 1995-12-23 1995-12-23 A voice recognition device for car KR100262564B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019950055190A KR100262564B1 (en) 1995-12-23 1995-12-23 A voice recognition device for car

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019950055190A KR100262564B1 (en) 1995-12-23 1995-12-23 A voice recognition device for car

Publications (2)

Publication Number Publication Date
KR970037890A true KR970037890A (en) 1997-07-24
KR100262564B1 KR100262564B1 (en) 2000-08-01

Family

ID=19443621

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019950055190A KR100262564B1 (en) 1995-12-23 1995-12-23 A voice recognition device for car

Country Status (1)

Country Link
KR (1) KR100262564B1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR19980023857A (en) * 1996-09-30 1998-07-06 배순훈 Car Audio Noise Reduction Device for Car Voice Recognition System
KR20000033530A (en) * 1998-11-24 2000-06-15 김영환 Car noise removing method using voice section detection and spectrum subtraction

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR19980023857A (en) * 1996-09-30 1998-07-06 배순훈 Car Audio Noise Reduction Device for Car Voice Recognition System
KR20000033530A (en) * 1998-11-24 2000-06-15 김영환 Car noise removing method using voice section detection and spectrum subtraction

Also Published As

Publication number Publication date
KR100262564B1 (en) 2000-08-01

Similar Documents

Publication Publication Date Title
EP0763811B1 (en) Speech signal processing apparatus for detecting a speech signal
JPH0990974A (en) Signal processor
DE68929442D1 (en) Arrangement for determining the presence of speech sounds
CN1110790C (en) Controller for starting vehicle by means of phoneme
KR910020645A (en) Voice signal coding device
KR910020644A (en) Voice Noise Separator
KR910015962A (en) Voice signal processing device
KR910020643A (en) Voice signal processing device
EP0614169B1 (en) Voice signal processing device
JP2004341033A (en) Voice mediated activating unit and its method
KR970037890A (en) Car Speech Recognition Device
JP3523382B2 (en) Voice recognition device and voice recognition method
JP2992324B2 (en) Voice section detection method
JP3114757B2 (en) Voice recognition device
KR100246617B1 (en) Speech detection method using the continuous pitch information
JP2580768B2 (en) Voice recognition device
JPS6336000B2 (en)
JPS63278100A (en) Voice recognition equipment
JPH03160499A (en) Speech recognizing device
KR100334238B1 (en) Apparatus and method for detecting speech/non-speech using the envelope of speech waveform
KR900016943A (en) Voice recognition device
JPS6227798A (en) Voice recognition equipment
JPS62150299A (en) Voice signal section detector
KR940018736A (en) Pseudo-voice Separator and Method of Speech Recognition System
JPS63226692A (en) Pattern comparison system

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E701 Decision to grant or registration of patent right
GRNT Written decision to grant
FPAY Annual fee payment

Payment date: 20060503

Year of fee payment: 7

LAPS Lapse due to unpaid annual fee