Publication Details

Further Progress in Meeting Recognition: The ICSI-SRI Spring 2005 Speech-to-Text Evaluation System

STOLCKE Andreas, ANGUERA Xavier, BOAKYE Kofi, CETIN Özgür, GRÉZL František, JANIN Adam, MANDAL Arindam, PESKIN Barbara, WOOTERS Chuck and ZHENG Jing. Further Progress in Meeting Recognition: The ICSI-SRI Spring 2005 Speech-to-Text Evaluation System. In: Machine Learning for Multimodal Interaction, Second International Workshop, MLMI 2005, Edinburgh, UK, July 11-13, 2005, Revised Selected Papers. Lecture Notes in Computer Science 3869, Springer 2006. Edinburgh, Scotland: University of Edinburgh, 2005, pp. 463-475. ISBN 978-3-540-32549-9.
Czech title
Další postup v rozpoznávání schůzí: ICSI-SRI jaro2005 řeč-text evaluační systém
Type
conference paper
Language
english
Authors
Stolcke Andreas (SRI)
Anguera Xavier (UPC)
Boakye Kofi (ICSI Berkeley)
Cetin Özgür (ICSI Berkeley)
Grézl František, Ing., Ph.D. (DCGM FIT BUT)
Janin Adam (ICSI Berkeley)
Mandal Arindam (UWASH)
Peskin Barbara (ICSI Berkeley)
Wooters Chuck (ICSI Berkeley)
Zheng Jing (SRI)
Keywords

speech-to-text system, rich transcription, system adaptation

Abstract

Development of the speech-to-text system for meeting transcription. Adaptation of the system from conversational telephone speech domain to meeting domain.

Annotation

We describe the development of our speech recognition system for
the NIST Spring 2005 Meeting Rich Transcription (RT-05S) evaluation, highlighting improvements made since last year. The system is based on the SRIICSI-UW RT-04F conversational telephone speech (CTS) recognition system, with meeting-adapted models and various audio preprocessing steps. This year's system features better delay-sum processing of distant microphone channels and energy-based crosstalk suppression for close-talking microphones. Acoustic modeling is improved by virtue of various enhancements to the background (CTS) models, including added training data, decision-tree based state tying, and the inclusion of discriminatively trained phone posterior features estimated by multilayer perceptrons. In particular, we make use of adaptation of both acoustic models and MLP features to the meeting domain. For distant microphone recognition we obtained considerable gains by combining and cross-adapting narrowband (telephone) acousticmodels with broadband (broadcast news) models. Language models (LMs) were improved with the inclusion of new meeting and web data. In spite of a lack of training data, we created effective LMs for the CHIL lecture domain. Results are reported on RT-04S and RT-05S meeting data. Measured on RT-04S conference data, we achieved an overall improvement of 17% relative in bothMDM and IHMconditions compared to last year's evaluation system. Results on lecture data are comparable to the best reported results for that task.

Published
2005
Pages
463-475
Proceedings
Machine Learning for Multimodal Interaction, Second International Workshop, MLMI 2005, Edinburgh, UK, July 11-13, 2005, Revised Selected Papers
Series
Lecture Notes in Computer Science 3869, Springer 2006
Conference
2nd Joint Workshop on Multimodal Interaction and Related Machine Learning Algorithms, Edinburgh, GB
ISBN
978-3-540-32549-9
Publisher
University of Edinburgh
Place
Edinburgh, Scotland, GB
BibTeX 
@INPROCEEDINGS{FITPUB7910,
   author = "Andreas Stolcke and Xavier Anguera and Kofi Boakye and {\"{O}}zg{\"{u}}r Cetin and Franti\v{s}ek Gr\'{e}zl and Adam Janin and Arindam Mandal and Barbara Peskin and Chuck Wooters and Jing Zheng",
   title = "Further Progress in Meeting Recognition: The ICSI-SRI Spring 2005 Speech-to-Text Evaluation System",
   pages = "463--475",
   booktitle = "Machine Learning for Multimodal Interaction, Second International Workshop, MLMI 2005, Edinburgh, UK, July 11-13, 2005, Revised Selected Papers",
   series = "Lecture Notes in Computer Science 3869, Springer 2006",
   year = 2005,
   location = "Edinburgh, Scotland, GB",
   publisher = "University of Edinburgh",
   ISBN = "978-3-540-32549-9",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/7910"
}
Back to top