A per-layer learnable threshold, trained jointly with CTC fine-tuning, prunes 60-65% of parameters from wav2vec2-base and HuBERT-large without a statistically significant WER increase on test-clean.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Effective and Efficient One-pass Compression of Speech Foundation Models Using Sparsity-aware Self-pinching Gates
A per-layer learnable threshold, trained jointly with CTC fine-tuning, prunes 60-65% of parameters from wav2vec2-base and HuBERT-large without a statistically significant WER increase on test-clean.