JPS6480999A - Voice initial end detector - Google Patents

Voice initial end detector

Info

Publication number
JPS6480999A
JPS6480999A JP62238451A JP23845187A JPS6480999A JP S6480999 A JPS6480999 A JP S6480999A JP 62238451 A JP62238451 A JP 62238451A JP 23845187 A JP23845187 A JP 23845187A JP S6480999 A JPS6480999 A JP S6480999A
Authority
JP
Japan
Prior art keywords
consonant
start end
section
voice
misrecognition
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
JP62238451A
Other languages
Japanese (ja)
Inventor
Shigeaki Komatsu
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Brother Industries Ltd
Original Assignee
Brother Industries Ltd
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 Brother Industries Ltd filed Critical Brother Industries Ltd
Priority to JP62238451A priority Critical patent/JPS6480999A/en
Publication of JPS6480999A publication Critical patent/JPS6480999A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE: To prevent the generation of misrecognition by detecting the start end of a consonant part of a voice in relation to calculated energy >= about 700Hz. CONSTITUTION: An amplifier 12 amplifies the level of a voice collected by a microphone 11 to a level matched with succeeding processing. A RAM 17 is used for working an intermediate/high frequency energy change rate buffer 17e in which the short time energy change rate of a frequency component >=700Hz is written. Then a CPU 15 detects a retrieving section, i.e., a vowel section, for detecting a frame to be the start end of a consonant and detects the start end of the consonant by setting up a frame other than the vowel section as a consonant detection section. Consequently error-reduced detection is accomplished without being affected by a buzz sound or the like, and in the case of executing voice recognition by using the start end of the detected consonant, misrecognition is suppressed.
JP62238451A 1987-09-22 1987-09-22 Voice initial end detector Pending JPS6480999A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP62238451A JPS6480999A (en) 1987-09-22 1987-09-22 Voice initial end detector

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP62238451A JPS6480999A (en) 1987-09-22 1987-09-22 Voice initial end detector

Publications (1)

Publication Number Publication Date
JPS6480999A true JPS6480999A (en) 1989-03-27

Family

ID=17030415

Family Applications (1)

Application Number Title Priority Date Filing Date
JP62238451A Pending JPS6480999A (en) 1987-09-22 1987-09-22 Voice initial end detector

Country Status (1)

Country Link
JP (1) JPS6480999A (en)

Similar Documents

Publication Publication Date Title
KR840003871A (en) Speech recognition method and device
KR880701435A (en) Voice processing device
JPS58130393A (en) Voice recognition equipment
EP0303022A3 (en) Rapidly training a speech recognizer to a subsequent speaker given training data of a reference speaker
ATE9415T1 (en) VOICE RECOGNITION SYSTEM.
Ericsdotter et al. Gender differences in vowel duration in read Swedish: Preliminary results
JPS6480999A (en) Voice initial end detector
Eukel Phonotactic basis for word frequency effects: Implications for lexical distance metrics
Miller et al. Vowel normalization: Differences between vowels spoken by children, women, and men
JPS5491009A (en) Audio recognition unit
Meinedo et al. Syllable onset detection applied to the portuguese language.
EP0173986A3 (en) Method of and device for the recognition, without previous training of connected words belonging to small vocabularies
AU1700788A (en) An adaptive threshold voiced detector
JPS54139307A (en) Pitch extraction unit
JPS6462700A (en) Voice recognition equipment
Aubert Supervised segmentation with application to speech recognition.
JPS6459394A (en) Digital voice extractor
Niita et al. Speaker-independent word recognition using multiple similarity method.
Miura Production and perception of English sentences spoken by Japanese university students.
JPS6454494A (en) Voice segmentation apparatus
Kasuya An improved autocorrelation pitch detector
Seki et al. Factors affecting voicing distinction of stops for the hearing impaired.
JPS5923398A (en) Mono-syllable voice recognition equipment
Boyanov et al. Robust pitch detection for normal and pathologic voice
Ma et al. Robust speech feature extraction based on dynamic minimum subband spectral subtraction