AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Speech-T: Transducer for Text to Speech and Beyond

Jiawei Chen, Xu Tan, Yichong Leng, Jin Xu, Guihua Wen, Tao Qin, Tie-Yan Liu

NeurIPS 2021 · Architectures and efficiency › Attention and transformers · approach: Efficient language transformers · 5 citations (OpenAlex)

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
use Transducer with monotonic alignment instead of attention alignmentTTS · 50 hard sentenceserror rate: 32% -> 0%Improved
use Speech-T instead of TransformerTTS for voice qualityTTS · LJSpeechMOS: 3.77 ± 0.07 -> 3.74 ± 0.07No clear effect
increase look-ahead k in streaming encoder (ablation)Speech-T · Streaming TTSMOS: 3.14 ± 0.19 -> 3.44 ± 0.15Improved
allow k=2 look-ahead words versus k=0 (ablation)Speech-T · Streaming TTSMOS: 3.14 ± 0.19 -> 3.68 ± 0.13Improved
use lazy forward algorithm instead of GMM-based forward algorithm (ablation)Speech-T · LJSpeechCMOS: -1.54 -> 0Improved
use lazy forward algorithm instead of default Transducer forward algorithm (ablation)Speech-T · LJSpeechCMOS: N/A -> 0Improved
add proposed alignment constraint (ablation)Speech-T · LJSpeechCMOS: N/A -> 0Improved
use proposed alignment constraint instead of handcrafted diagonal constraint (ablation)Speech-T · LJSpeechCMOS: -1.33 -> 0Improved
unify TTS and ASR in one model instead of separate TTS modelSpeech-T streaming · LJSpeechCMOS: -0.09 -> 0No clear effect
unify TTS and ASR in one model instead of separate ASR modelSpeech-T streaming · LJSpeechPER: 9.7% -> 9.0%Improved

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