Adding a noise classification head to Whisper during fine-tuning improved call-for-help detection accuracy from 65% to 88% on real-world recordings, though out-of-domain noise accuracy remained low.
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
-
Noise-Agnostic Multitask Whisper Training for Reducing False Alarm Errors in Call-for-Help Detection
Adding a noise classification head to Whisper during fine-tuning improved call-for-help detection accuracy from 65% to 88% on real-world recordings, though out-of-domain noise accuracy remained low.