16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

Unsupervised Adaptation for Deep Neural Network Using Linear Least Square Method

Roger Hsiao, Tim Ng, Stavros Tsakalidis, Long Nguyen, Richard Schwartz

Raytheon BBN Technologies, USA

In this paper, we propose a novel model based adaptation for deep neural networks based on a linear least square method. Our proposed algorithm can perform unsupervised adaptation even if the auto transcripts may have 60-70% of word error rate. We evaluate our algorithm on low resource languages, from the IARPA BABEL program, such as Assamese, Bengali, Haitian Creole, Lao and Zulu. Our experiments focus on unsupervised speaker, dialect and environment adaptation and we show that it can improve both speech recognition and keyword search performance.

Full Paper

Bibliographic reference.  Hsiao, Roger / Ng, Tim / Tsakalidis, Stavros / Nguyen, Long / Schwartz, Richard (2015): "Unsupervised adaptation for deep neural network using linear least square method", In INTERSPEECH-2015, 2887-2891.