Search papers, labs, and topics across Lattice.
3
0
6
5
Reinforcement learning outperforms supervised fine-tuning in adapting ASR systems to synthetic speech, achieving a 40% reduction in word error rates.
Just 4 hours of speech data closes the modality gap in LLM-based ASR, rivaling full-dataset fine-tuning and unlocking effective domain adaptation.
LLM-based ASR can get a context boost without the compute cost: compress prior audio turns into learned latent tokens and retain transcripts to recover accuracy while shrinking the audio footprint.
