A framework using native-only trained discrete token surprisal and DTW alignment features improves pronunciation assessment PCC to 0.66 on SpeechOcean762, approaching supervised performance.
Exploring self-supervised audio models for generalized anomalous sound detection,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Light-weight Pronunciation Assessment via Discrete Speech Token Surprisal
A framework using native-only trained discrete token surprisal and DTW alignment features improves pronunciation assessment PCC to 0.66 on SpeechOcean762, approaching supervised performance.