Timezone: »

Attention-Based Models for Speech Recognition
Jan K Chorowski · Dzmitry Bahdanau · Dmitriy Serdyuk · Kyunghyun Cho · Yoshua Bengio

Wed Dec 09 04:00 PM -- 08:59 PM (PST) @ 210 C #5

Recurrent sequence generators conditioned on input data through an attention mechanism have recently shown very good performance on a range of tasks including machine translation, handwriting synthesis and image caption generation. We extend the attention-mechanism with features needed for speech recognition. We show that while an adaptation of the model used for machine translation reaches a competitive 18.6\% phoneme error rate (PER) on the TIMIT phoneme recognition task, it can only be applied to utterances which are roughly as long as the ones it was trained on. We offer a qualitative explanation of this failure and propose a novel and generic method of adding location-awareness to the attention mechanism to alleviate this issue. The new method yields a model that is robust to long inputs and achieves 18\% PER in single utterances and 20\% in 10-times longer (repeated) utterances. Finally, we propose a change to the attention mechanism that prevents it from concentrating too much on single frames, which further reduces PER to 17.6\% level.

Author Information

Jan K Chorowski (University of Wroclaw)
Dzmitry Bahdanau (Jacobs University, Germany)
Dmitriy Serdyuk (Université de Montréal)
Kyunghyun Cho (NYU)

Kyunghyun Cho is an associate professor of computer science and data science at New York University and a research scientist at Facebook AI Research. He was a postdoctoral fellow at the Université de Montréal until summer 2015 under the supervision of Prof. Yoshua Bengio, and received PhD and MSc degrees from Aalto University early 2014 under the supervision of Prof. Juha Karhunen, Dr. Tapani Raiko and Dr. Alexander Ilin. He tries his best to find a balance among machine learning, natural language processing, and life, but almost always fails to do so.

Yoshua Bengio (U. Montreal)

More from the Same Authors