Research map / Large language models / Language, safety and interpretability
StableToken: A Noise-Robust Semantic Speech Tokenizer for Resilient SpeechLLMs
Yuhan Song, Linhao Zhang, Chuhan Wu, Aiwei Liu, Wei Jia, Houfeng Wang, Zhou Xiao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use StableToken instead of best supervised tokenizer | supervised semantic tokenizer · FLEURS noisy avg | UED: 26.17 -> 10.17 | Improved |
| add consensus loss (ablation) | StableToken · FLEURS Real (OOD) | UED: 17.43 -> 10.96 | Improved |
| add noise-aware training (ablation) | StableToken · LibriSpeech-other | WER: 5.52 -> 4.88 | Improved |
| add multi-branch voting architecture (ablation) | StableToken · FLEURS Real (OOD) | UED: 24.47 -> 21.51 | Improved |
| increase voters from 3 to 5 (ablation) | StableToken · FLEURS Gaussian noise | UED: 20.66 -> 18.68 | Improved |
| increase voters from 5 to 7 (ablation) | StableToken · FLEURS Gaussian noise | UED: 18.68 -> 18.10 | No clear effect |
| use StableToken tokenizer in SpeechLLM instead of GLM-4-Voice | Qwen2.5-3B SpeechLLM · CHiME-4 test real | WER: 51.08 -> 35.90 | Improved |
| use StableToken tokenizer in SpeechLLM instead of GLM-4-Voice | Qwen2.5-3B SpeechLLM · SEED-TTS EN | MOS: 4.19 -> 4.12 | Mixed |
| place quantizer at layer 24 instead of 16 (ablation) | StableToken · Gaussian UED vs ESD SER | UED / SER acc: UED 18.68 -> 13.65; SER 67.38 -> 59.51 | Mixed |
| place quantizer at layer 8 instead of 16 (ablation) | StableToken · Gaussian UED | UED: 18.68 -> 22.05 | Hurt |
| use 4:1 clean-to-noisy branch ratio instead of 3:2 (ablation) | StableToken · Gaussian UED | UED: 18.68 -> 20.94 | Hurt |
Closest papers by meaning
- SpeechTokenizer: Unified Speech Tokenizer for Speech Language Models (ICLR 2024 · similarity 0.87)
- WavTokenizer: an Efficient Acoustic Discrete Codec Tokenizer for Audio Language Modeling (ICLR 2025 · similarity 0.87)
- ALMTokenizer: A Low-bitrate and Semantic-rich Audio Codec Tokenizer for Audio Language Modeling (ICML 2025 · similarity 0.87)
- What Was That Again? Certified Robustness for Automatic Speech Recognition (NeurIPS 2026 · similarity 0.86)
- CLaM-TTS: Improving Neural Codec Language Model for Zero-Shot Text-to-Speech (ICLR 2024 · similarity 0.85)
- Generative Audio Language Modeling with Continuous-valued Tokens and Masked Next-Token Prediction (ICML 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
