Speaker-Disentangled Chunk-Wise Regression for Syllabic Tokenization
A speaker-disentangled syllabic tokenizer regresses perturbed student representations toward clean teacher targets to improve syllable boundary detection and speech language modeli…
Hugging Face · Daily Papers
·Ryota Komatsu, Kota Kawakita
·
Este artigo está em destaque na seleção diária de papers do Hugging Face, curada pela comunidade de pesquisa em IA.
Autores: Ryota Komatsu, Kota Kawakita, Takuma Okamoto, Takahiro Shinozaki
- 0 upvotes da comunidade
- Temas: syllabic tokenization, teacher-student distillation, HuBERT, latent speech frame representations, speaker-disentangled, cross-entropy objective
Resumo
Resumo original (em inglês), extraído do paper:
A speaker-disentangled syllabic tokenizer regresses perturbed student representations toward clean teacher targets to improve syllable boundary detection and speech language modeling performance.