In speech recognition systeme dealing with unlimited vocabulary and based on stochastic language models, when the target recognition task is changed, recognition performance decreases because the language model is no longer appropriate. This paper describes two approaches for adapting a specific/general syllable trigram model to a new task. One uses a amall amount of text data similar to the target task, and the other uses supervised learning using the most recent input phrases and similar text. In this paper, these adaptation methods are called
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Sho-ichi MATSUNAGA, Tomokazu YAMADA, Kiyohiro SHIKANO, "Task Adaptation in Syllable Trigram Models for Continuous Speech Recognition" in IEICE TRANSACTIONS on Information,
vol. E76-D, no. 1, pp. 38-43, January 1993, doi: .
Abstract: In speech recognition systeme dealing with unlimited vocabulary and based on stochastic language models, when the target recognition task is changed, recognition performance decreases because the language model is no longer appropriate. This paper describes two approaches for adapting a specific/general syllable trigram model to a new task. One uses a amall amount of text data similar to the target task, and the other uses supervised learning using the most recent input phrases and similar text. In this paper, these adaptation methods are called
URL: https://global.ieice.org/en_transactions/information/10.1587/e76-d_1_38/_p
Copy
@ARTICLE{e76-d_1_38,
author={Sho-ichi MATSUNAGA, Tomokazu YAMADA, Kiyohiro SHIKANO, },
journal={IEICE TRANSACTIONS on Information},
title={Task Adaptation in Syllable Trigram Models for Continuous Speech Recognition},
year={1993},
volume={E76-D},
number={1},
pages={38-43},
abstract={In speech recognition systeme dealing with unlimited vocabulary and based on stochastic language models, when the target recognition task is changed, recognition performance decreases because the language model is no longer appropriate. This paper describes two approaches for adapting a specific/general syllable trigram model to a new task. One uses a amall amount of text data similar to the target task, and the other uses supervised learning using the most recent input phrases and similar text. In this paper, these adaptation methods are called
keywords={},
doi={},
ISSN={},
month={January},}
Copy
TY - JOUR
TI - Task Adaptation in Syllable Trigram Models for Continuous Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 38
EP - 43
AU - Sho-ichi MATSUNAGA
AU - Tomokazu YAMADA
AU - Kiyohiro SHIKANO
PY - 1993
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E76-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 1993
AB - In speech recognition systeme dealing with unlimited vocabulary and based on stochastic language models, when the target recognition task is changed, recognition performance decreases because the language model is no longer appropriate. This paper describes two approaches for adapting a specific/general syllable trigram model to a new task. One uses a amall amount of text data similar to the target task, and the other uses supervised learning using the most recent input phrases and similar text. In this paper, these adaptation methods are called
ER -