Comparing SMT methods for automatic generation of pronunciation variants

Authors:
Panagiota Karanasou;Lori Lamel
Affiliations:
LIMSI, CNRS, Orsay, France;LIMSI, CNRS, Orsay, France
Venue:
IceTAL'10 Proceedings of the 7th international conference on Advances in natural language processing
Year:
2010

Citing 9
Cited 3

Automatic generation of multiple pronunciations based on neural networks

Speech Communication
Pronunciation variants across system configuration, language and speaking style

Speech Communication - Special issue on modeling pronunciation variation for automatic speech recognition
Information Retrieval

Information Retrieval
Algorithms for grapheme-phoneme translation for English and French: applications for database searches and speech synthesis

Computational Linguistics
BLEU: a method for automatic evaluation of machine translation

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Paraphrasing with bilingual parallel corpora

ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics
Moses: open source toolkit for statistical machine translation

ACL '07 Proceedings of the 45th Annual Meeting of the ACL on Interactive Poster and Demonstration Sessions
Coping with out-of-vocabulary words: Open versus huge vocabulary asr

ICASSP '09 Proceedings of the 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
Pronunciation Modeling With Reduced Confusion for Mandarin Chinese Using a Three-Stage Framework

IEEE Transactions on Audio, Speech, and Language Processing

Automatic speech recognition for under-resourced languages: A survey

Speech Communication
Web-based tools and methods for rapid pronunciation dictionary creation

Speech Communication
SMT-based ASR domain adaptation methods for under-resourced languages: Application to Romanian

Speech Communication

Quantified Score

Hi-index	0.00

Visualization

Abstract

Multiple-pronunciation dictionaries are often used by automatic speech recognition systems in order to account for different speaking styles. In this paper, two methods based on statistical machine translation (SMT) are used to generate multiple pronunciations from the canonical pronunciation of a word. In the first method, a machine translation tool is used to perform phoneme-to-phoneme (p2p) conversion and derive variants from a given canonical pronunciation. The second method is based on a pivot method proposed for the paraphrase extraction task. The two methods are compared under different training conditions which allow single or multiple pronunciations in the training set, and their performance is evaluated in terms of recall and precision measures.