Speech Tokenizers are Vulnerable: Transferable Semantic Attack and Robust Tokenizer
eess.AS, cs.AI, cs.CR, cs.SD
Submitted: 2026-11-11
Updated: 2026-11-11
Project page: https://wxzyd123.github.io/t-semattack
Terminology
Sources
- Qwen2-Audio Technical Report
- Qwen3-Omni Technical Report
- Tacotron: Towards End-to-End Speech Synthesis
- Audio Flamingo 3: Advancing Audio Intelligence with Fully Open Large Audio Language Models
- Qwen3-ASR Technical Report
- StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
- SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
- Discrete Audio Tokens: More Than a Survey!
- CosyVoice: A Scalable Multilingual Zero-shot Text-to-speech Synthesizer based on Supervised Semantic Tokens
- CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis
- LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech
- Seed-TTS: A Family of High-Quality Versatile Speech Generation Models
- VIBEVOICE-ASR Technical Report
- FireRedASR: Open-Source Industrial-Grade Mandarin Speech Recognition Models from Encoder-Decoder to LLM Integration
- Fun-ASR Technical Report
- The Interspeech 2026 Audio Encoder Capability Challenge for Large Audio Language Models
- DashengTokenizer: One layer is enough for unified audio understanding and generation
- Step-Audio: Unified Understanding and Generation in Intelligent Speech Interaction
Related papers
- WavCube: Unifying Speech Representation for Understanding and Generation via Semantic-Acoustic Joint Modeling
- From perception to production: how acoustic invariance facilitates articulatory learning in a self-supervised vocal imitation model
- Read What You Hear: Reference-Free Hypotheses Evaluation with Acoustic Discrepancy
- pyFDN: A Python Toolbox for Feedback Delay Network Design, Analysis, and Optimization
- GigaSpeechBench: A Real-World Multilingual Speech-to-Text Benchmark
- DM-ASR: Diarization-aware Multi-speaker ASR with Unified Offline and Online Decoding with Large Language Models