TW279969B - Voice synthesis method of chinese with finite word templates - Google Patents

Voice synthesis method of chinese with finite word templates

Info

Publication number
TW279969B
TW279969B TW83104254A TW83104254A TW279969B TW 279969 B TW279969 B TW 279969B TW 83104254 A TW83104254 A TW 83104254A TW 83104254 A TW83104254 A TW 83104254A TW 279969 B TW279969 B TW 279969B
Authority
TW
Taiwan
Prior art keywords
group
unit
word
voice synthesis
synthesis unit
Prior art date
Application number
TW83104254A
Other languages
Chinese (zh)
Inventor
Jiunn-Yuh Tsay
Jih-Shyh Liou
Wenn-Shing Lay
Jiunn-Sheau Lii
Wenn-Jiunn Wang
Pei-Yi Ding
Original Assignee
Telecomm Lab Dgt Motc
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 Telecomm Lab Dgt Motc filed Critical Telecomm Lab Dgt Motc
Priority to TW83104254A priority Critical patent/TW279969B/en
Application granted granted Critical
Publication of TW279969B publication Critical patent/TW279969B/en

Links

Abstract

A voice synthesis method of Chinese with finite word templatescomprises five steps of: building selective database of voice synthesis unit by digitally recording all possible combinations of finite word templates, the segmenting the recorded data with word template as unit to accommodate into selective database of voice synthesis unit; group analysis by forming one parent group with set of same word template from selective database of voice synthesis unit. Each word template of parent group is individually put into phrase-head,phrase-mid and phrase-tail three groups according to the wordtemplate position in combination phrase of finite word template; extracting feature parameter by calculating linear prediction coding coefficient of each frame for data in selective database of voice synthesis unit, then calculating each frame's LPC-derived spectrum up to order 13; deciding representative synthesis unit in group by calculating sum of distance score of candidate unit between each other in group. The unit with lowest sum of distance score is representative unit of the group; text-speech of infinite word from finite word template by converting input word into combination of group code, then concatenating synthesis unit representing the group directly to synthesis speech.
TW83104254A 1994-05-11 1994-05-11 Voice synthesis method of chinese with finite word templates TW279969B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW83104254A TW279969B (en) 1994-05-11 1994-05-11 Voice synthesis method of chinese with finite word templates

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW83104254A TW279969B (en) 1994-05-11 1994-05-11 Voice synthesis method of chinese with finite word templates

Publications (1)

Publication Number Publication Date
TW279969B true TW279969B (en) 1996-07-01

Family

ID=51397551

Family Applications (1)

Application Number Title Priority Date Filing Date
TW83104254A TW279969B (en) 1994-05-11 1994-05-11 Voice synthesis method of chinese with finite word templates

Country Status (1)

Country Link
TW (1) TW279969B (en)

Similar Documents

Publication Publication Date Title
Bulyko et al. A bootstrapping approach to automating prosodic annotation for limited-domain synthesis
Hirst Form and function in the representation of speech prosody
CN1889170B (en) Method and system for generating synthesized speech based on recorded speech template
US7266495B1 (en) Method and system for learning linguistically valid word pronunciations from acoustic data
Nakajima et al. Automatic generation of synthesis units based on context oriented clustering
EP0953970A3 (en) Method and apparatus using decision trees to generate and score multiple pronunciations for a spelled word
EP0847041A3 (en) Method and apparatus for speech recognition performing noise adaptation
EP1038292A4 (en) System and method for auditorially representing pages of sgml data
EP0831460A3 (en) Speech synthesis method utilizing auxiliary information
EP2506252A3 (en) Topic specific models for text formatting and speech recognition
DE69033651D1 (en) Method and device with bit allocation for the quantization and dequantization of transformed speech signals.
ATE15415T1 (en) METHOD AND DEVICE FOR REDUNDANCY-REDUCING DIGITAL SPEECH PROCESSING.
CN108763338A (en) A kind of News Collection&Edit System based on power industry
CN1300049A (en) Method and apparatus for identifying speech sound of chinese language common speech
CN1333501A (en) Dynamic Chinese speech synthesizing method
EP0798695A3 (en) Speech recognizing method and apparatus
TW279969B (en) Voice synthesis method of chinese with finite word templates
KR20020031305A (en) Method for the encoding of prosody for a speech encoder working at very low bit rates
Kumar et al. Significance of durational knowledge for speech synthesis system in an Indian language
SE9601812L (en) Enhancements to, or with respect to, speech-to-speech conversion
Furui Speech recognition-past, present, and future
Hirst ProZed: a multilingual prosody editor for speech synthesis
Yoshimura et al. Pitch pattern clustering of user utterances in human-machine dialogue
JPH08212190A (en) Support device for production of multimedia data
JP3009709B2 (en) Japanese speech recognition method