AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Scaling Transformers for End-to-End Discrete Audio Tokenization

Yitian Gong, Kuangwei Chen, Zhaoye Fei, Xiaogui Yang, Ke Chen, Yang Wang, Kexin Huang, Mingshu Chen, Ruixiao Li, Qinyuan Cheng, Shimin Li, Xipeng Qiu

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: Efficient language transformers

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 changedWhereMeasuredEffect
add Progressive Sequence Dropout (p>0) during AR TTS training (ablation)TAC-TTS · Seed-TTS-Eval, reduced bitrateSIM and WERImproved
vary dropout probability p among 0.25, 0.5, 1.0 (ablation)TAC-TTS · Seed-TTS-EvalSIM and WERNo clear effect
increase dropout probability p (ablation)TAC-TTS · Seed-TTS-Evaltraining GPU memoryImproved
train without Progressive Sequence Dropout (p=0.0) (ablation)TAC-TTS · Seed-TTS-Eval, full bitrateSIM and WERNo clear effect
full end-to-end optimization instead of freezing encoder and quantizer (ablation)TAC · speech reconstruction, 500k stepsSIM, STOI, PESQImproved
freeze encoder and quantizer (partial optimization) (ablation)TAC · speech reconstructionSIM, STOI, PESQHurt
increase parameters from 319M to 1169M (ablation)TAC · speech reconstruction, 0.5-4 kbpsSIM, STOI, PESQImproved
scale parameters without raising bitrate (1169M at low bitrate) (ablation)TAC · speech reconstruction, low bitrateSIM, STOI, PESQNo clear effect
raise bitrate from 750 to 4000 bps (ablation)TAC · speech reconstruction SIM (EN)SIM: 0.82 -> 0.97Improved
use fully Transformer, end-to-end tokenizer vs hybrid baselines at 1000 bpsTAC · speech reconstruction SIM (EN)SIM: 0.85 -> 0.88Improved
use TAC tokens for AR TTS vs prior discrete AR TTSTAC-TTS · Seed-TTS-Eval ENSIM: 67.7 -> 73.1Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy