tensorflow_end2end_speech_recognition icon indicating copy to clipboard operation
tensorflow_end2end_speech_recognition copied to clipboard

Published 20 hours ago • verified publisher icon hirofumi0810

Location-Based Attention Layer

Open Peidong-Wang opened this issue 6 years ago • 0 comments

It seems that attention layer type 'location' should actually be 'hybrid' [1,2].

[1] Kim, Suyoun, Takaaki Hori, and Shinji Watanabe. "Joint CTC-attention based end-to-end speech recognition using multi-task learning." Acoustics, Speech and Signal Processing (ICASSP), 2017 IEEE International Conference on. IEEE, 2017. [2] Bahdanau, Dzmitry, et al. "End-to-end attention-based large vocabulary speech recognition." Acoustics, Speech and Signal Processing (ICASSP), 2016 IEEE International Conference on. IEEE, 2016.

Peidong-Wang avatar Feb 15 '18 01:02 Peidong-Wang