Weakly-supervised text-to-speech alignment confidence measure

Guillaume Serrière, Christophe Cerisara, Dominique Fohr, Odile Mella


Abstract
This work proposes a new confidence measure for evaluating text-to-speech alignment systems outputs, which is a key component for many applications, such as semi-automatic corpus anonymization, lips syncing, film dubbing, corpus preparation for speech synthesis and speech recognition acoustic models training. This confidence measure exploits deep neural networks that are trained on large corpora without direct supervision. It is evaluated on an open-source spontaneous speech corpus and outperforms a confidence score derived from a state-of-the-art text-to-speech aligner. We further show that this confidence measure can be used to fine-tune the output of this aligner and improve the quality of the resulting alignment.
Anthology ID:
C16-1192
Volume:
Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers
Month:
December
Year:
2016
Address:
Osaka, Japan
Editors:
Yuji Matsumoto, Rashmi Prasad
Venue:
COLING
SIG:
Publisher:
The COLING 2016 Organizing Committee
Note:
Pages:
2042–2050
Language:
URL:
https://aclanthology.org/C16-1192
DOI:
Bibkey:
Cite (ACL):
Guillaume Serrière, Christophe Cerisara, Dominique Fohr, and Odile Mella. 2016. Weakly-supervised text-to-speech alignment confidence measure. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 2042–2050, Osaka, Japan. The COLING 2016 Organizing Committee.
Cite (Informal):
Weakly-supervised text-to-speech alignment confidence measure (Serrière et al., COLING 2016)
Copy Citation:
PDF:
https://aclanthology.org/C16-1192.pdf