An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation
碩士 === 國立交通大學 === 電信工程系 === 88 === In this thesis, we implement a way be able to cut waves automatically to do preprocessing of Mandarin text-to-speech system. We choice HMM sub-syllable models which are used in speech recognition. In the beginning, we use the speech database recorded by NTU, NCTU,...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2000
|
Online Access: | http://ndltd.ncl.edu.tw/handle/14190620692907064065 |
id |
ndltd-TW-088NCTU0435042 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-088NCTU04350422016-07-08T04:22:39Z http://ndltd.ncl.edu.tw/handle/14190620692907064065 An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation 使用語音辨認做前處理之TTS系統發展 Wei-Chih Kuo 郭威志 碩士 國立交通大學 電信工程系 88 In this thesis, we implement a way be able to cut waves automatically to do preprocessing of Mandarin text-to-speech system. We choice HMM sub-syllable models which are used in speech recognition. In the beginning, we use the speech database recorded by NTU, NCTU, and NCKU. About the state observation probability, we employ mixture Gaussian models, and raise the numbers of Gaussian distribution. In additional, we employ SBR to compensate the effect of speakers and channels. Final, we get a series of HMM sub-syllable models which make the recognition rate about 70%. We employ the models to cut the speech database of a single female speaker, and extract the prosodic features from the cutting position. Then, we use the prosodic features to retrain the prosodic parameters by RNN prosodic generator. Final, we adopt the prosodic parameters to implement a female Mandarin text-to-speech system, and the syllable energy contour is taken as a prosodic information. The female Mandarin text-to-speech system consists of four main parts: text analyzer, RNN prosodic generator, waveform inventory of synthesis units, and PSOLA synthesizer. Sin-Horng Chen 陳信宏 2000 學位論文 ; thesis 49 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立交通大學 === 電信工程系 === 88 === In this thesis, we implement a way be able to cut waves automatically to do preprocessing of Mandarin text-to-speech system. We choice HMM sub-syllable models which are used in speech recognition. In the beginning, we use the speech database recorded by NTU, NCTU, and NCKU. About the state observation probability, we employ mixture Gaussian models, and raise the numbers of Gaussian distribution. In additional, we employ SBR to compensate the effect of speakers and channels. Final, we get a series of HMM sub-syllable models which make the recognition rate about 70%.
We employ the models to cut the speech database of a single female speaker, and extract the prosodic features from the cutting position. Then, we use the prosodic features to retrain the prosodic parameters by RNN prosodic generator. Final, we adopt the prosodic parameters to implement a female Mandarin text-to-speech system, and the syllable energy contour is taken as a prosodic information. The female Mandarin text-to-speech system consists of four main parts: text analyzer, RNN prosodic generator, waveform inventory of synthesis units, and PSOLA synthesizer.
|
author2 |
Sin-Horng Chen |
author_facet |
Sin-Horng Chen Wei-Chih Kuo 郭威志 |
author |
Wei-Chih Kuo 郭威志 |
spellingShingle |
Wei-Chih Kuo 郭威志 An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
author_sort |
Wei-Chih Kuo |
title |
An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
title_short |
An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
title_full |
An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
title_fullStr |
An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
title_full_unstemmed |
An Implementation of Mandarin TTS System using Preprocessing based on HMM segmentation |
title_sort |
implementation of mandarin tts system using preprocessing based on hmm segmentation |
publishDate |
2000 |
url |
http://ndltd.ncl.edu.tw/handle/14190620692907064065 |
work_keys_str_mv |
AT weichihkuo animplementationofmandarinttssystemusingpreprocessingbasedonhmmsegmentation AT guōwēizhì animplementationofmandarinttssystemusingpreprocessingbasedonhmmsegmentation AT weichihkuo shǐyòngyǔyīnbiànrènzuòqiánchùlǐzhīttsxìtǒngfāzhǎn AT guōwēizhì shǐyòngyǔyīnbiànrènzuòqiánchùlǐzhīttsxìtǒngfāzhǎn AT weichihkuo implementationofmandarinttssystemusingpreprocessingbasedonhmmsegmentation AT guōwēizhì implementationofmandarinttssystemusingpreprocessingbasedonhmmsegmentation |
_version_ |
1718339444868644864 |