Incremental unsupervised training for university lecture recognition

Michael Heck, Sebastian Stüker, Sakriani Sakti, Alex Waibel, Satoshi Nakamura


Abstract
In this paper we describe our work on unsupervised adaptation of the acoustic model of our simultaneous lecture translation system. We trained a speaker independent acoustic model, with which we produce automatic transcriptions of new lectures in order to improve the system for a specific lecturer. We compare our results against a model that was trained in a supervised way on an exact manual transcription. We examine four different ways of processing the decoder outputs of the automatic transcription with respect to the treatment of pronunciation variants and noise words. We will show that, instead of fixating the latter informations in the transcriptions, it is of advantage to let the Viterbi algorithm during training decide which pronunciations to use and where to insert which noise words. Further, we utilize word level posterior probabilities obtained during decoding by weighting and thresholding the words of a transcription.
Anthology ID:
2013.iwslt-papers.8
Volume:
Proceedings of the 10th International Workshop on Spoken Language Translation: Papers
Month:
December 5-6
Year:
2013
Address:
Heidelberg, Germany
Venue:
IWSLT
SIG:
SIGSLT
Publisher:
Note:
Pages:
Language:
URL:
https://aclanthology.org/2013.iwslt-papers.8
DOI:
Bibkey:
Cite (ACL):
Michael Heck, Sebastian Stüker, Sakriani Sakti, Alex Waibel, and Satoshi Nakamura. 2013. Incremental unsupervised training for university lecture recognition. In Proceedings of the 10th International Workshop on Spoken Language Translation: Papers, Heidelberg, Germany.
Cite (Informal):
Incremental unsupervised training for university lecture recognition (Heck et al., IWSLT 2013)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2013.iwslt-papers.8.pdf