JPS6481999A - Phoneme string conversion system - Google Patents

Phoneme string conversion system

Info

Publication number
JPS6481999A
JPS6481999A JP24009487A JP24009487A JPS6481999A JP S6481999 A JPS6481999 A JP S6481999A JP 24009487 A JP24009487 A JP 24009487A JP 24009487 A JP24009487 A JP 24009487A JP S6481999 A JPS6481999 A JP S6481999A
Authority
JP
Japan
Prior art keywords
character string
neural network
phoneme
symbol sequence
supplied
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
JP24009487A
Other languages
Japanese (ja)
Inventor
Kazuhiko Ono
Katsunobu Fushikida
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.)
NEC Corp
Original Assignee
NEC Corp
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 NEC Corp filed Critical NEC Corp
Priority to JP24009487A priority Critical patent/JPS6481999A/en
Publication of JPS6481999A publication Critical patent/JPS6481999A/en
Pending legal-status Critical Current

Links

Abstract

PURPOSE: To improve the recognition rate of a character string by learning the regularity of errors generated by a sound processing part by using the learning function of a neural network. CONSTITUTION: A speech waveform 8 is analyzed by the sound processing part 9 to obtain a phoneme symbol sequence 10, which contains no error, so a pair of the phoneme symbol sequence and a correct answer character string are supplied to the neural network 11 and learnt. The neural network learns the mutual relation between the phoneme symbol sequence and correct answer character string to gradually acquire a function for estimating a proper character string from the phoneme symbols. At the time of speech/character string conversion, the phoneme symbols are supplied to the neural network to start firing, and a character string 12 is obtained as the stimulation output of an output layer. Consequently, the performance of word recognition is improved.
JP24009487A 1987-09-24 1987-09-24 Phoneme string conversion system Pending JPS6481999A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP24009487A JPS6481999A (en) 1987-09-24 1987-09-24 Phoneme string conversion system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP24009487A JPS6481999A (en) 1987-09-24 1987-09-24 Phoneme string conversion system

Publications (1)

Publication Number Publication Date
JPS6481999A true JPS6481999A (en) 1989-03-28

Family

ID=17054400

Family Applications (1)

Application Number Title Priority Date Filing Date
JP24009487A Pending JPS6481999A (en) 1987-09-24 1987-09-24 Phoneme string conversion system

Country Status (1)

Country Link
JP (1) JPS6481999A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0371196A (en) * 1989-08-10 1991-03-26 Yamaha Corp Method for generating musical sound parameter
JPH06274197A (en) * 1993-03-18 1994-09-30 Sharp Corp Speech recognition device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0371196A (en) * 1989-08-10 1991-03-26 Yamaha Corp Method for generating musical sound parameter
JPH06274197A (en) * 1993-03-18 1994-09-30 Sharp Corp Speech recognition device

Similar Documents

Publication Publication Date Title
Rabiner et al. Isolated and connected word recognition-theory and selected applications
McClelland et al. The TRACE model of speech perception
EP0313975A3 (en) Design and construction of a binary-tree system for language modelling
GB2303237A (en) Method of training neural networks used for speech recognition
FR2520912B1 (en) CONTINUOUS SPEECH RECOGNITION METHOD AND APPARATUS
AU637144B2 (en) Pattern recognition device
EP0303022A3 (en) Rapidly training a speech recognizer to a subsequent speaker given training data of a reference speaker
EP0758781A3 (en) Utterance verification using word based minimum verification error training for recognizing a keyword string
EP0241768A3 (en) Synthesizing word baseforms used in speech recognition
DE3884880D1 (en) Cheap speech recognition device and method.
DE3275779D1 (en) Recognition of speech or speech-like sounds
JPS53114601A (en) Voice recognition system
GB2304507A (en) Speech-recognition system utilizing neural networks and method of using same
JPS6481999A (en) Phoneme string conversion system
Bobrow et al. A limited speech recognition system
CN1009320B (en) Speech recognition
Di Martino Dynamic time warping algorithms for isolated and connected word recognition
DE3680903D1 (en) METHOD FOR GENERATING PRE WORD PATTERNS FROM SEVERAL EXPRESSIONS FOR VOICE RECOGNITION.
DE3378384D1 (en) Recognition of continuous speech
JPS6453259A (en) Sentence proofreading device
Ainsworth Audio feedback for error correction in a digit recognition task.
JPS6437599A (en) Japanese sentence voice input device
JPS6428699A (en) Continuous voice recognition system
SU639535A1 (en) Method of correcting deformation of crus
Scarr Speech recognition by machineߞart or science?