Signals 4 · free daily AI digest

EmoRES-TTS: Residual-Enhanced Vector Steering for Emotional Speech Generation

Paper recorded by Signals 4 on 2026-09-29 in cs.CL. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-29 on arXiv · recorded by Signals 4 on 2026-09-30

Category: cs.CL · 自然语言处理 · first seen 2026-09-30

Abstract

Emotion-conditioned text-to-speech (TTS) models may fail to express the requested emotion reliably, and improving controllability by additional training is costly in both computation and emotion-labeled speech training data. We therefore study vector steering, a training-free approach that modifies the internal representations of a frozen model. CoCoEmo, a conventional vector steering method for e

Read on arXiv →

#2 most recent of 292 cs.CL papers we have recorded · ↑ newer: Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering · ↓ older: Pretraining Latent Information Feedback Transformers with Teacher Supe
Cite this page: EmoRES-TTS: Residual-Enhanced Vector Steering for Emotional Speech Generation: the #2 most recent of 292 cs.CL papers we have recorded (as of 2026-09-29). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/emores-tts-residual-enhanced-vector-steering-for-emotional-speech-generation.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.CL papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions