Loud and Clear: Dynamic Activation Steering for Improving Speech Intelligibility in Noisy Environments

arXiv:2610.07647 · cs.SD, cs.CL · Submitted 2026-10-06 · Read on arXiv

cs.SD, cs.CL

Submitted: 2026-10-06

Updated: 2026-10-06

Project page: https://seymanurakti.github.io/loud-and-clear

Terminology

• Activation Steering — 1× in this paper · explained in 4 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• Voice Conversion for Lombard Speaking Style with Implicit and Explicit Acoustic Feature Conditioning
• Analysis and Synthesis of Hypo and Hyperarticulated Speech
• Synthesizing the Lombard Effect: Multi-Level Control of Speech Clarity and Vocal Effort in TTS
• Steering Language Models With Activation Engineering
• CoCoEmo: Composable and Controllable Human-Like Emotional TTS via Activation Steering
• EmoSteer-TTS: Fine-Grained and Training-Free Emotion-Controllable Text-to-Speech via Activation Steering
• Activation Steering for Accent-Neutralized Zero-Shot Text-To-Speech
• Prompt-Activation Duality: Improving Activation Steering via Attention-Level Interventions
• Qwen3-TTS Technical Report
• EARS: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberation

Related papers

• Few-Shot Open-Set Audio Classification via Transductive Prototype Refinement and Class Logit Enhancement
• Spectral Masking and Interpolation Attack (SMIA): A Black-box Adversarial Attack against Voice Authentication and Anti-Spoofing Systems
• Remix the Timbre: Diffusion-Based Style Transfer Across Polyphonic Stems
• AVMeme Exam: A Multimodal Multilingual Multicultural Benchmark for LLMs' Contextual and Cultural Knowledge and Thinking
• SoundWeaver: Compositional Warm-Starting for Text-to-Audio Diffusion Serving
• WASIL: In-the-Wild Arabic Spoken Interactions with LLMs