An Improved Model for Voicing Silent Speech

David Gaddy, Dan Klein


Abstract
In this paper, we present an improved model for voicing silent speech, where audio is synthesized from facial electromyography (EMG) signals. To give our model greater flexibility to learn its own input features, we directly use EMG signals as input in the place of hand-designed features used by prior work. Our model uses convolutional layers to extract features from the signals and Transformer layers to propagate information across longer distances. To provide better signal for learning, we also introduce an auxiliary task of predicting phoneme labels in addition to predicting speech audio features. On an open vocabulary intelligibility evaluation, our model improves the state of the art for this task by an absolute 25.8%.
Anthology ID:
2021.acl-short.23
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
Month:
August
Year:
2021
Address:
Online
Editors:
Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
175–181
Language:
URL:
https://aclanthology.org/2021.acl-short.23
DOI:
10.18653/v1/2021.acl-short.23
Bibkey:
Cite (ACL):
David Gaddy and Dan Klein. 2021. An Improved Model for Voicing Silent Speech. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 175–181, Online. Association for Computational Linguistics.
Cite (Informal):
An Improved Model for Voicing Silent Speech (Gaddy & Klein, ACL-IJCNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-short.23.pdf
Video:
 https://aclanthology.org/2021.acl-short.23.mp4
Code
 dgaddy/silent_speech