In this paper, we propose a PLSA-based language model for sports-related live speech. This model is implemented using a unigram rescaling technique that combines a topic model and an n-gram. In the conventional method, unigram rescaling is performed with a topic distribution estimated from a recognized transcription history. This method can improve the performance, but it cannot express topic transition. By incorporating the concept of topic transition, it is expected that the recognition performance will be improved. Thus, the proposed method employs a "Topic HMM" instead of a history to estimate the topic distribution. The Topic HMM is an Ergodic HMM that expresses typical topic distributions as well as topic transition probabilities. Word accuracy results from our experiments confirmed the superiority of the proposed method over a trigram and a PLSA-based conventional method that uses a recognized history.
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
Atsushi SAKO, Tetsuya TAKIGUCHI, Yasuo ARIKI, "Language Modeling Using PLSA-Based Topic HMM" in IEICE TRANSACTIONS on Information,
vol. E91-D, no. 3, pp. 522-528, March 2008, doi: 10.1093/ietisy/e91-d.3.522.
Abstract: In this paper, we propose a PLSA-based language model for sports-related live speech. This model is implemented using a unigram rescaling technique that combines a topic model and an n-gram. In the conventional method, unigram rescaling is performed with a topic distribution estimated from a recognized transcription history. This method can improve the performance, but it cannot express topic transition. By incorporating the concept of topic transition, it is expected that the recognition performance will be improved. Thus, the proposed method employs a "Topic HMM" instead of a history to estimate the topic distribution. The Topic HMM is an Ergodic HMM that expresses typical topic distributions as well as topic transition probabilities. Word accuracy results from our experiments confirmed the superiority of the proposed method over a trigram and a PLSA-based conventional method that uses a recognized history.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e91-d.3.522/_p
Copy
@ARTICLE{e91-d_3_522,
author={Atsushi SAKO, Tetsuya TAKIGUCHI, Yasuo ARIKI, },
journal={IEICE TRANSACTIONS on Information},
title={Language Modeling Using PLSA-Based Topic HMM},
year={2008},
volume={E91-D},
number={3},
pages={522-528},
abstract={In this paper, we propose a PLSA-based language model for sports-related live speech. This model is implemented using a unigram rescaling technique that combines a topic model and an n-gram. In the conventional method, unigram rescaling is performed with a topic distribution estimated from a recognized transcription history. This method can improve the performance, but it cannot express topic transition. By incorporating the concept of topic transition, it is expected that the recognition performance will be improved. Thus, the proposed method employs a "Topic HMM" instead of a history to estimate the topic distribution. The Topic HMM is an Ergodic HMM that expresses typical topic distributions as well as topic transition probabilities. Word accuracy results from our experiments confirmed the superiority of the proposed method over a trigram and a PLSA-based conventional method that uses a recognized history.},
keywords={},
doi={10.1093/ietisy/e91-d.3.522},
ISSN={1745-1361},
month={March},}
Copy
TY - JOUR
TI - Language Modeling Using PLSA-Based Topic HMM
T2 - IEICE TRANSACTIONS on Information
SP - 522
EP - 528
AU - Atsushi SAKO
AU - Tetsuya TAKIGUCHI
AU - Yasuo ARIKI
PY - 2008
DO - 10.1093/ietisy/e91-d.3.522
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E91-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2008
AB - In this paper, we propose a PLSA-based language model for sports-related live speech. This model is implemented using a unigram rescaling technique that combines a topic model and an n-gram. In the conventional method, unigram rescaling is performed with a topic distribution estimated from a recognized transcription history. This method can improve the performance, but it cannot express topic transition. By incorporating the concept of topic transition, it is expected that the recognition performance will be improved. Thus, the proposed method employs a "Topic HMM" instead of a history to estimate the topic distribution. The Topic HMM is an Ergodic HMM that expresses typical topic distributions as well as topic transition probabilities. Word accuracy results from our experiments confirmed the superiority of the proposed method over a trigram and a PLSA-based conventional method that uses a recognized history.
ER -