Back to feed
arXiv cs.CL
arXiv cs.CL
6/23/2026
EmoInstruct-TTS: Dual-Path Instruction-Guided Emotional Speech Synthesis

EmoInstruct-TTS: Dual-Path Instruction-Guided Emotional Speech Synthesis

Short summary

EmoInstruct-TTS proposes a dual-path framework for emotional speech synthesis that handles 48 fine-grained emotional states via natural language instructions. An Instruction-Conditioned Emotion Flow Model (ICE-Flow) converts free-form directives into emotion embeddings integrated with LLM-based synthesis pipelines. The approach delivers improved emotional controllability and speech naturalness over existing baselines.

  • •Dual-path framework supporting 48 fine-grained emotional states (intensity + nuance)
  • •ICE-Flow model infers emotion embeddings from free-form natural language instructions
  • •LLM-integrated synthesis pipeline preserves semantic planning while enabling explicit emotional control

Generated with AI, which can make mistakes.

#research-breakthrough#ai-tools#ai-agents
Read full article at arXiv cs.CL

Is this a good recommendation for you?

Explore more

Continuous Audio Thinking for Large Audio Language Models
arXiv cs.CL
MLLP-VRAIN UPV system for the IWSLT 2026 Simultaneous Speech Translation task
arXiv cs.CL
Graph-Based Phonetic Error Correction of Noisy ASR
arXiv cs.CL
Are you speaking my languages? On spoken language adherence in multimodal LLMs
arXiv cs.CL