An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface

O. Walter, V. Despotovic, R. Haeb-Umbach, J. Gemmeke, B. Ons, H. Van hamme, in: INTERSPEECH 2014, 2014.

Conference Paper | English
Author
; ; ; ; ;
Abstract
In this paper, we investigate unsupervised acoustic model training approaches for dysarthric-speech recognition. These models are first, frame-based Gaussian posteriorgrams, obtained from Vector Quantization (VQ), second, so-called Acoustic Unit Descriptors (AUDs), which are hidden Markov models of phone-like units, that are trained in an unsupervised fashion, and, third, posteriorgrams computed on the AUDs. Experiments were carried out on a database collected from a home automation task and containing nine speakers, of which seven are considered to utter dysarthric speech. All unsupervised modeling approaches delivered significantly better recognition rates than a speaker-independent phoneme recognition baseline, showing the suitability of unsupervised acoustic model training for dysarthric speech. While the AUD models led to the most compact representation of an utterance for the subsequent semantic inference stage, posteriorgram-based representations resulted in higher recognition rates, with the Gaussian posteriorgram achieving the highest slot filling F-score of 97.02%. Index Terms: unsupervised learning, acoustic unit descriptors, dysarthric speech, non-negative matrix factorization
Publishing Year
Proceedings Title
INTERSPEECH 2014
LibreCat-ID

Cite this

Walter O, Despotovic V, Haeb-Umbach R, Gemmeke J, Ons B, Van hamme H. An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface. In: INTERSPEECH 2014. ; 2014.
Walter, O., Despotovic, V., Haeb-Umbach, R., Gemmeke, J., Ons, B., & Van hamme, H. (2014). An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface. In INTERSPEECH 2014.
@inproceedings{Walter_Despotovic_Haeb-Umbach_Gemmeke_Ons_Van hamme_2014, title={An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface}, booktitle={INTERSPEECH 2014}, author={Walter, Oliver and Despotovic, Vladimir and Haeb-Umbach, Reinhold and Gemmeke, Jrt and Ons, Bart and Van hamme, Hugo}, year={2014} }
Walter, Oliver, Vladimir Despotovic, Reinhold Haeb-Umbach, Jrt Gemmeke, Bart Ons, and Hugo Van hamme. “An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface.” In INTERSPEECH 2014, 2014.
O. Walter, V. Despotovic, R. Haeb-Umbach, J. Gemmeke, B. Ons, and H. Van hamme, “An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface,” in INTERSPEECH 2014, 2014.
Walter, Oliver, et al. “An Evaluation of Unsupervised Acoustic Model Training for a Dysarthric Speech Interface.” INTERSPEECH 2014, 2014.

Link(s) to Main File(s)
Access Level
Restricted Closed Access
External material:
Supplementary Material
Description
Poster
Supplementary Material
Description
Spotlight

Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar