The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm.
碩士 === 國立中山大學 === 電機工程學系研究所 === 91 === Based on Hidden Markov Models (HMM) with One-Stage Dynamic Programming Algorithm, a continuous-speech and speaker-independent Mandarin digit speech recognition system was designed in this work. In order to implement this architecture to fit the performance of h...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2003
|
Online Access: | http://ndltd.ncl.edu.tw/handle/00457077193899947281 |
id |
ndltd-TW-091NSYS5442041 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-091NSYS54420412016-06-22T04:20:47Z http://ndltd.ncl.edu.tw/handle/00457077193899947281 The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. 結合隱藏式馬可夫模型與一階動態規劃演算法之連續語音辨識系統. Fang-Yi Hsieh 謝芳易 碩士 國立中山大學 電機工程學系研究所 91 Based on Hidden Markov Models (HMM) with One-Stage Dynamic Programming Algorithm, a continuous-speech and speaker-independent Mandarin digit speech recognition system was designed in this work. In order to implement this architecture to fit the performance of hardware, various parameters of speech characteristics were defined to optimize the process. Finally, the “State Duration” and the “Tone Transition Property Parameter” were extracted from speech temporal information to improve the recognition rate. Via using the test database, experimental results show that this new ideal of one-stage dynamic programming algorithm , with “state duration” and “ tone transition property parameter” , will have 18% recognition rate increase when compare to the conventional one. For speaker-independent and connect-word recognition, this system will achieve recognition rate to 74%. For speaker-independent but isolate-word recognition, it will have recognition rate higher than 96%. Recognition rate of 92% is obtained as this system is applied to the connect-word speaker-dependent recognition. Chern Tzuen-Lih 陳遵立 2003 學位論文 ; thesis 97 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中山大學 === 電機工程學系研究所 === 91 === Based on Hidden Markov Models (HMM) with One-Stage Dynamic Programming Algorithm, a continuous-speech and speaker-independent Mandarin digit speech recognition system was designed in this work.
In order to implement this architecture to fit the performance of hardware, various parameters of speech characteristics were defined to optimize the process. Finally, the “State Duration” and the “Tone Transition Property Parameter” were extracted from speech temporal information to improve the recognition rate.
Via using the test database, experimental results show that this new ideal of one-stage dynamic programming algorithm , with “state duration” and “ tone transition property parameter” , will have 18% recognition rate increase when compare to the conventional one. For speaker-independent and connect-word recognition, this system will achieve recognition rate to 74%. For speaker-independent but isolate-word recognition, it will have recognition rate higher than 96%. Recognition rate of 92% is obtained as this system is applied to the connect-word speaker-dependent recognition.
|
author2 |
Chern Tzuen-Lih |
author_facet |
Chern Tzuen-Lih Fang-Yi Hsieh 謝芳易 |
author |
Fang-Yi Hsieh 謝芳易 |
spellingShingle |
Fang-Yi Hsieh 謝芳易 The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
author_sort |
Fang-Yi Hsieh |
title |
The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
title_short |
The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
title_full |
The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
title_fullStr |
The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
title_full_unstemmed |
The Continuous Speech Recognition System Base on Hidden Markov Models with One-Stage Dynamic Programming Algorithm. |
title_sort |
continuous speech recognition system base on hidden markov models with one-stage dynamic programming algorithm. |
publishDate |
2003 |
url |
http://ndltd.ncl.edu.tw/handle/00457077193899947281 |
work_keys_str_mv |
AT fangyihsieh thecontinuousspeechrecognitionsystembaseonhiddenmarkovmodelswithonestagedynamicprogrammingalgorithm AT xièfāngyì thecontinuousspeechrecognitionsystembaseonhiddenmarkovmodelswithonestagedynamicprogrammingalgorithm AT fangyihsieh jiéhéyǐncángshìmǎkěfūmóxíngyǔyījiēdòngtàiguīhuàyǎnsuànfǎzhīliánxùyǔyīnbiànshíxìtǒng AT xièfāngyì jiéhéyǐncángshìmǎkěfūmóxíngyǔyījiēdòngtàiguīhuàyǎnsuànfǎzhīliánxùyǔyīnbiànshíxìtǒng AT fangyihsieh continuousspeechrecognitionsystembaseonhiddenmarkovmodelswithonestagedynamicprogrammingalgorithm AT xièfāngyì continuousspeechrecognitionsystembaseonhiddenmarkovmodelswithonestagedynamicprogrammingalgorithm |
_version_ |
1718318532021714944 |