[go: up one dir, main page]

TW324821B - Speech recognition method and apparatus for specific talker - Google Patents

Speech recognition method and apparatus for specific talker

Info

Publication number
TW324821B
TW324821B TW083108581A TW83108581A TW324821B TW 324821 B TW324821 B TW 324821B TW 083108581 A TW083108581 A TW 083108581A TW 83108581 A TW83108581 A TW 83108581A TW 324821 B TW324821 B TW 324821B
Authority
TW
Taiwan
Prior art keywords
speech
tested
reference sample
close
input
Prior art date
Application number
TW083108581A
Other languages
Chinese (zh)
Inventor
Yuan-Kwei Suen
Ruey-Shyang Luo
Chyau-Ian Day
Long-Jang Chen
Original Assignee
United Microelectronics 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 United Microelectronics Corp filed Critical United Microelectronics Corp
Priority to TW083108581A priority Critical patent/TW324821B/en
Application granted granted Critical
Publication of TW324821B publication Critical patent/TW324821B/en

Links

Landscapes

  • Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)

Abstract

A speech recognition method and apparatus for specific talker, used to compare one input tested speech and multiple pre-stored reference sample, consists of the steps: (1) Transforming input tested speech signal into one series of original speech parameter frame; (2) Using the above original speech parameter frame to decide start/end point of input speech, then according to start/end point of input speech deciding sound spectrum parameter frame series; (3) Using one nonlinear time domain normalization method transforming the above sound spectrum parameter frame series to speech feature parameter frame series with predetermined length; (4) Applying amplitude quantization normalization to the above speech feature parameter; (5) Comparing speech feature parameter of tested speech and speech feature parameter of all reference samples, and finding out reference sample most close to tested speech and reference sample next to most close tested speech; (6) According to reference sample most close to tested speech and reference sample next to most close tested speech, comparing its difference with one standard threshold, thereby deciding recognition result.
TW083108581A 1994-09-16 1994-09-16 Speech recognition method and apparatus for specific talker TW324821B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW083108581A TW324821B (en) 1994-09-16 1994-09-16 Speech recognition method and apparatus for specific talker

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW083108581A TW324821B (en) 1994-09-16 1994-09-16 Speech recognition method and apparatus for specific talker

Publications (1)

Publication Number Publication Date
TW324821B true TW324821B (en) 1998-01-11

Family

ID=58262116

Family Applications (1)

Application Number Title Priority Date Filing Date
TW083108581A TW324821B (en) 1994-09-16 1994-09-16 Speech recognition method and apparatus for specific talker

Country Status (1)

Country Link
TW (1) TW324821B (en)

Similar Documents

Publication Publication Date Title
MY104270A (en) Method and apparatus for real time speech recognition with and without speaker dependency
Daubechies et al. A nonlinear squeezing of the continuous wavelet transform based on auditory nerve models
CA1059631A (en) Method of judging voiced and unvoiced conditions of speech signal
Schroeder Reference signal for signal quality studies
US4181813A (en) System and method for speech recognition
FI20145179A7 (en) Method and apparatus for multi-level distributed speech recognition
DE69725172D1 (en) METHOD AND DEVICE FOR DETECTING NOISE SAMPLE SAMPLES FROM A NOISE
TW340938B (en) Dialog device with voice identification capabilty
MX9505296A (en) Speech recognition bias equalization method and apparatus.
EP0788091A3 (en) Speech encoding and decoding method and apparatus therefor
KR950034055A (en) Digitalized Speech Signal Analysis Method for Excitation Parameter Determination and Speech Encoding System
ATE127303T1 (en) METHOD FOR IDENTIFYING VOICE AND CALL HISTORY SIGNALS.
KR890002816A (en) Cheap speech recognition system and method
AU2001277647A1 (en) Method for noise robust classification in speech coding
EP0854469A3 (en) Speech encoding apparatus and method
EP0862162A3 (en) Speech recognition using nonparametric speech models
CA2154881A1 (en) A system and method for compression and decompression of audio signals
EP0798695A3 (en) Speech recognizing method and apparatus
BR9709010B1 (en) coin testing method and apparatus, use of a coin testing apparatus, and coin recognition process.
DE69739545D1 (en) METHOD AND SYSTEM FOR THE AUTOMATIC TEXT-INDEPENDENT EVALUATION OF THE LANGUAGE DIRECTORY
TW324821B (en) Speech recognition method and apparatus for specific talker
US3603738A (en) Time-domain pitch detector and circuits for extracting a signal representative of pitch-pulse spacing regularity in a speech wave
DE59609067D1 (en) Method for the automatic recognition of a spoken text
ATE250314T1 (en) DEVICE AND METHOD FOR ANALYZING THE SPECTRAL REPRESENTATION OF A DECODED TIME SIGNALS
DE3279549D1 (en) Apparatus and method for articulatory speech recognition

Legal Events

Date Code Title Description
MK4A Expiration of patent term of an invention patent