Speech synthesis using HMM technique
This dissertation implements the speech recognition of letters of alphabet and digits with and accuracy of 96.15% and 100% respectively and also implements and HMM-based speech synthesis system in which the speech waveform is generated from HMMs themselves. The system is modeled by multispace probab...
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis |
Language: | English |
Published: |
2009
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/18814 |
_version_ | 1811682115257892864 |
---|---|
author | May, Thwe Khaing. |
author2 | Foo Say Wei |
author_facet | Foo Say Wei May, Thwe Khaing. |
author_sort | May, Thwe Khaing. |
collection | NTU |
description | This dissertation implements the speech recognition of letters of alphabet and digits with and accuracy of 96.15% and 100% respectively and also implements and HMM-based speech synthesis system in which the speech waveform is generated from HMMs themselves. The system is modeled by multispace probability distribution HMMs and multi-dimensional Gaussian distributions respectively. The distributions for spectral parameter, pitch parameter and the state duration are clustered independently by using a decision-tree based vocoding technique. The proposed system has been confirmed successfully that it synthesized natural-sounding speech which resembles the speaker in the training database, this hidden Markov Model (HMM) has found widespread use in automatic speech recognition. And the system can change voice qualities of synthesized speech by transforming HMM parameters. |
first_indexed | 2024-10-01T03:51:42Z |
format | Thesis |
id | ntu-10356/18814 |
institution | Nanyang Technological University |
language | English |
last_indexed | 2024-10-01T03:51:42Z |
publishDate | 2009 |
record_format | dspace |
spelling | ntu-10356/188142023-07-04T15:21:23Z Speech synthesis using HMM technique May, Thwe Khaing. Foo Say Wei School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing This dissertation implements the speech recognition of letters of alphabet and digits with and accuracy of 96.15% and 100% respectively and also implements and HMM-based speech synthesis system in which the speech waveform is generated from HMMs themselves. The system is modeled by multispace probability distribution HMMs and multi-dimensional Gaussian distributions respectively. The distributions for spectral parameter, pitch parameter and the state duration are clustered independently by using a decision-tree based vocoding technique. The proposed system has been confirmed successfully that it synthesized natural-sounding speech which resembles the speaker in the training database, this hidden Markov Model (HMM) has found widespread use in automatic speech recognition. And the system can change voice qualities of synthesized speech by transforming HMM parameters. Master of Science (Signal Processing) 2009-07-20T03:04:12Z 2009-07-20T03:04:12Z 2008 2008 Thesis http://hdl.handle.net/10356/18814 en 86 p. application/pdf |
spellingShingle | DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing May, Thwe Khaing. Speech synthesis using HMM technique |
title | Speech synthesis using HMM technique |
title_full | Speech synthesis using HMM technique |
title_fullStr | Speech synthesis using HMM technique |
title_full_unstemmed | Speech synthesis using HMM technique |
title_short | Speech synthesis using HMM technique |
title_sort | speech synthesis using hmm technique |
topic | DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing |
url | http://hdl.handle.net/10356/18814 |
work_keys_str_mv | AT maythwekhaing speechsynthesisusinghmmtechnique |