%Aigaion2 BibTeX export from Idiap Publications
%Saturday 17 October 2026 04:00:09 PM

@INPROCEEDINGS{Kumar_SLT2026_2026,
                      author = {Kumar, Shashi and Labrak, Yanis and Watawana, Hasindri Sankalpana and Burdisso, Sergio and Villatoro-Tello, Esa{\'{u}} and Hacioğlu, Kadri and Motlicek, Petr and Stolcke, Andreas},
                    projects = {UNIPHORE, ELOQUENCE},
         mainresearchprogram = {AI for Everyone},
  additionalresearchprograms = {AI for Life},
                       title = {When Synthetic Speech Is All You Have: Better Call GRPO},
                   booktitle = {IEEE Spoken Language Technology 2026},
                        year = {2026},
                         url = {https://arxiv.org/abs/2607.08409},
                    abstract = {LLM-based ASR adapted to regulated domains such as banking is bottlenecked by privacy: real speech is costly and legally constrained to collect, making synthetic text-to-speech (TTS) an attractive substitute. Yet synthetic speech stays acoustically mismatched with real recordings, and work on this gap has stayed within supervised fine-tuning (SFT). We instead turn to reinforcement learning, and show that Group Relative Policy Optimization (GRPO) extracts far more from the same synthetic speech than SFT. Synthetic-only adaptation of the model with GRPO, a critic-free method rewarding low-WER hypotheses, reduces WER by 40\% relative to SFT (36.71\%→22.09\%), and an SFT-then-GRPO combination pushes this further to 45\%. We trace the gain to behavior rather than representation: GRPO reduces insertion errors by improving stopping calibration and speech-to-text alignment by better anchoring attention to audio, leaving early-layer representations intact. When synthetic speech is the main resource, reinforcement learning should be preferred over supervised fine-tuning.},
                         pdf = {https://publications.idiap.ch/attachments/papers/2026/Kumar_SLT2026_2026.pdf}
}