← 研踪 PaperTrailarXiv:2611.03900 · v1
Clean-anchored excess-emission control for CTC adaptation under competing speech
首次提交 2026-11-03 04:09
版本更新 2026-11-03 04:09
雷达观测 2026-11-04 23:40
摘要
Competing speech can cause excess emissions in a connectionist temporal classification (CTC) recognizer. We penalize the amount by which the noisy student's expected collapsed token count exceeds a frozen teacher's count on the paired clean utterance. The one-sided objective caches one scalar per utterance and adds no inference module. With 147,456 trainable parameters in wav2vec 2.0, it improves competing-speech word error rate (WER) over matched CTC-only adaptation across three development seeds. On a locked LibriSpeech test, speech-interference WER falls from 45.198% to 44.323%, while clean WER changes by 0.004 percentage points. Unseen VCTK interferers yield a 0.307-point improvement. Ablations favor collapsed-count control over frame-mass regularization. Longer training increases the speech gain but worsens noise and music performance, identifying an insertion-deletion tradeoff rather than uniform robustness.
观测到的版本
相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-03 04:09 · 最新观测版本
Clean-anchored excess-emission control for CTC adaptation under competing speech
Competing speech can cause excess emissions in a connectionist temporal classification (CTC) recognizer. We penalize the amount by which the noisy student's expected collapsed token count exceeds a frozen teacher's count on the paired clean utterance. The one-sided objective caches one scalar per utterance and adds no inference module. With 147,456 trainable parameters in wav2vec 2.0, it improves competing-speech word error rate (WER) over matched CTC-only adaptation across three development seeds. On a locked LibriSpeech test, speech-interference WER falls from 45.198% to 44.323%, while clean WER changes by 0.004 percentage points. Unseen VCTK interferers yield a 0.307-point improvement. Ablations favor collapsed-count control over frame-mass regularization. Longer training increases the speech gain but worsens noise and music performance, identifying an insertion-deletion tradeoff rather than uniform robustness.
查看此版本解读 →查看此版本原文 ↗