An Initial Study on Minimum Phone Error Discriminative Training for Continuous Phone Recognition System
碩士 === 國立清華大學 === 資訊工程學系 === 94 === Maximum Likelihood Estimation (MLE) is a traditional method for training acoustic models for speech recognition. This method does not consider discriminative relation between acoustic models, so some models are apt to obscure each other. In order to raise the diff...
Main Authors: | Shuo-Pin Hsu, 許碩斌 |
---|---|
Other Authors: | Jyh-Shing Roger Jang |
Format: | Others |
Language: | zh-TW |
Published: |
2006
|
Online Access: | http://ndltd.ncl.edu.tw/handle/05601427143243772117 |
Similar Items
-
An Initial Study on Minimum Phone Error Discriminative Learning of Acoustic Models for Mandarin Large Vocabulary Continuous Speech Recognition
by: Jen-Wei Kuo, et al.
Published: (2005) -
Minimum Phone Error Training for Code-Mixed Bilingual Speech Recognition
by: Hao-Chung Yang, et al.
Published: (2011) -
A Study on Minimum Phone Error Discriminative Training for Mandarin Chinese Speaker Adaptation
by: Niu Hsueh-Wen, et al.
Published: (2007) -
Minimum Phone Error Training of Acoustic Models and Features for Large Vocabulary Mandarin Speech Recognition
by: Jia-Yu Chen, et al.
Published: (2006) -
Evaluation and Analysis of Minimum Phone Error Training and Its Modified Versions for Large Vocabulary Mandarin Speech Recognition
by: Yung-Jen Cheng, et al.
Published: (2008)