On synthetic banking speech alone, GRPO cuts ASR WER 40% relative to SFT (36.71%→22.09%) by improving stopping calibration and attention anchoring to audio.
WavLM: Large-scale self-supervised pre-training for full stack speech processing
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Task-dependent simulation strategies for synthetic conversational data allow synthetic-only training to approach real-data baselines for multi-talker ASR and diarization, with mixing yielding further gains.
citing papers explorer
-
When Synthetic Speech Is All You Have: Better Call GRPO
On synthetic banking speech alone, GRPO cuts ASR WER 40% relative to SFT (36.71%→22.09%) by improving stopping calibration and attention anchoring to audio.
-
Mind the Gap: Impact of Synthetic Conversational Data on Multi-Talker ASR and Speaker Diarization
Task-dependent simulation strategies for synthetic conversational data allow synthetic-only training to approach real-data baselines for multi-talker ASR and diarization, with mixing yielding further gains.