My SLT 2018 Schedule

Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.

Create a login based on your email (takes less than one minute)
Perform 'Paper Search'
Select papers that you desire to save in your personalized schedule
Click on 'My Schedule' to see the current list of selected papers
Click on 'Printable Version' to create a separate window suitable for printing (the header and menu will appear, but will not actually print)

Paper Detail

Presentation #	15
Session:	ASR IV
Session Time:	Friday, December 21, 13:30 - 15:30
Presentation Time:	Friday, December 21, 13:30 - 15:30
Presentation:	Poster
Topic:	Speech recognition and synthesis:
Paper Title:	SPEAKER ADAPTATION FOR END-TO-END CTC MODELS
Authors:	Ke Li; Johns Hopkins University
	Jinyu Li; Microsoft AI and Research
	Yong Zhao; Microsoft AI and Research
	Kshitiz Kumar; Microsoft AI and Research
	Yifan Gong; Microsoft AI and Research
Abstract:	We propose two approaches for speaker adaptation in end-to-end (E2E) automatic speech recognition systems. One is Kullback-Leibler divergence (KLD) regularization and the other is multi-task learning (MTL). Both approaches aim to address the data sparsity especially output target sparsity issue of speaker adaptation in E2E systems. The KLD regularization adapts a model by forcing the output distribution from the adapted model to be close to the unadapted one. The MTL utilizes a jointly trained auxiliary task to improve the performance of the main task. We investigated our approaches on E2E connectionist temporal classification (CTC) models with three different types of output units. Experiments on the Microsoft short message dictation task demonstrated that MTL outperforms KLD regularization. In particular, the MTL adaptation obtained 8.8% and 4.0% relative word error rate reductions (WERRs) for supervised and unsupervised adaptations for the word CTC model, and produced 9.6% and 3.8% relative WERRs for the mix-unit CTC model, respectively.