An adaptive CFG method that tunes guidance based on LLM-detected mismatch between emotion prompts and text semantics improves emotional expressiveness in AR TTS while preserving audio quality and intelligibility.
Koel-tts: Enhancing llm based speech generation with preference alignment and classifier free guidance,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Mismatch Aware Guidance for Robust Emotion Control in Auto-Regressive TTS Models
An adaptive CFG method that tunes guidance based on LLM-detected mismatch between emotion prompts and text semantics improves emotional expressiveness in AR TTS while preserving audio quality and intelligibility.