Self-Supervised Syllable Discovery Based on Speaker-Disentangled HuBERT

[Paper] [Code] [Model]

Ryota Komatsu1, Kota Kawakita1, Takuma Okamoto2, Takahiro Shinozaki1

1Institute of Science Tokyo, 2National Institute of Information and Communications Technology

Model overview

Speech resynthesis samples from LibriSpeech test-clean

Original Resynthesis Syllable segmentation

License

The LibriSpeech dataset is made available under the CC BY 4.0 license.

References

  1. Vassil Panayotov, Guoguo Chen, Daniel Povey and Sanjeev Khudanpur, "LibriSpeech: an ASR corpus based on public domain audio books," in Proc. ICASSP, 2015, pp. 5206–5210.