Pith. sign in

These models usually operate over acoustic tokens or phonetic speech tokens (also known as semantic tokens)

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

background 1

citation-polarity summary

fields

cs.SD 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

support 1

representative citing papers

PAST: Phonetic-Acoustic Speech Tokenizer

cs.SD · 2025-05-20 · conditional · novelty 6.0

PAST jointly optimizes an EnCodec-style codec with CTC and phoneme classification losses to produce hybrid phonetic-acoustic speech tokens that outperform SpeechTokenizer and X-Codec.

citing papers explorer

Showing 1 of 1 citing paper.

  • PAST: Phonetic-Acoustic Speech Tokenizer cs.SD · 2025-05-20 · conditional · none · ref 1

    PAST jointly optimizes an EnCodec-style codec with CTC and phoneme classification losses to produce hybrid phonetic-acoustic speech tokens that outperform SpeechTokenizer and X-Codec.