Rohan Siva(@_rsiva):tts naturalness is hard. sft has its limits, but rl is tricky for flow-matching models since they lack the log-probs dpo and grpo rely on. excited to share work on adapting both methods for TTS with @cyrusasg at @DecagonAI!

Rohan Siva

@_rsiva

research @decagonai ml & robotics research @utaustin

加入 August 2025

29 正在关注 46 粉丝

Rohan Siva@_rsiva

2026.08.14 17:53

tts naturalness is hard. sft has its limits, but rl is tricky for flow-matching models since they lack the log-probs dpo and grpo rely on. excited to share work on adapting both methods for TTS with @cyrusasg at @DecagonAI!

显示更多

Decagon@DecagonAI

2026.08.14 17:21

Modern TTS models can sound great — and still fail badly on pacing, pauses, and prosody. We adapted DPO + GRPO to flow-matching models to tackle the tail end of TTS behavior:

显示更多