Language model parameter estimation using user transcriptions
Name
Bo-June-2009-Language model parameter estimation using user transcriptions.pdf
Size
117.28 KB
Format
Adobe PDF
Checksum (MD5)
1abefd849017eb747b2a7bc63e2b086c
Author(s)
Hsu, Bo-June
Glass, James R.
Date Issued
May 2009
Journal
Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, 2009
Publisher
Institute of Electrical and Electronics Engineers
Citation
Bo-June Hsu, and J. Glass. “Language model parameter estimation using user transcriptions.” Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on. 2009. 4805-4808. © 2009 IEEE
Version
Final published version
Abstract
In limited data domains, many effective language modeling techniques construct models with parameters to be estimated on an in-domain development set. However, in some domains, no such data exist beyond the unlabeled test corpus. In this work, we explore the iterative use of the recognition hypotheses for unsupervised parameter estimation. We also evaluate the effectiveness of supervised adaptation using varying amounts of user-provided transcripts of utterances selected via multiple strategies. While unsupervised adaptation obtains 80% of the potential error reductions, it is outperformed by using only 300 words of user transcription. By transcribing the lowest confidence utterances first, we further obtain an effective word error rate reduction of 0.6%.
Subjects
adaptation
language modeling
speech recognition
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1109/ICASSP.2009.4960706
