Projects per year
Abstract
The performance of automatic speech recognition systems can be improved by adapting an acoustic model to compensate for the mismatch between training and testing conditions, for example by adapting to unseen speakers. The success of speaker adaptation methods relies on selecting weights that are suitable for adaptation and using good adaptation schedules to update these weights in order not to overfit to the adaptation data. In this paper we investigate a principled way of adapting all the weights of the acoustic model using a meta-learning. We show that the meta-learner can learn to perform supervised and unsupervised speaker adaptation and that it outperforms a strong baseline adapting LHUC parameters when adapting a DNN AM with 1.5M parameters. We also report initial experiments on adapting TDNN AMs,
where the meta-learner achieves comparable performance with LHUC.
where the meta-learner achieves comparable performance with LHUC.
Original language | English |
---|---|
Title of host publication | Proc. of Interspeech 2018 |
Place of Publication | Hyderabad, India |
Publisher | ISCA |
Pages | 867-871 |
Number of pages | 5 |
DOIs | |
Publication status | Published - Sept 2018 |
Event | Interspeech 2018 - Hyderabad International Convention Centre, Hyderabad, India Duration: 2 Sept 2018 → 6 Sept 2018 http://interspeech2018.org/ |
Publication series
Name | |
---|---|
Publisher | ISCA |
ISSN (Electronic) | 1990-9772 |
Conference
Conference | Interspeech 2018 |
---|---|
Country/Territory | India |
City | Hyderabad |
Period | 2/09/18 → 6/09/18 |
Internet address |
Fingerprint
Dive into the research topics of 'Learning to Adapt: a Meta-learning Approach for Speaker Adaptation'. Together they form a unique fingerprint.Projects
- 1 Finished
-
SUMMA - Scalable Understanding of Mulitingual Media
Renals, S., Birch-Mayne, A. & Cohen, S.
1/02/16 → 31/01/19
Project: Research