Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Text-to-Speech Synthesis research benchmarks · SOTA2 Research

Text-to-Speech Synthesis

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

LibriTTS (CLEAN), LibriVox (NOISY), YouTube (WILD), and My Science Tutor (KIDS) (test)

3.7MOS

21

Mar 4, 2026

4.48MOS

11

May 29, 2026

4.63MOS

9

May 29, 2026

10-second speech segments

1.4Inference Time (s)

8

Feb 26, 2026

Internal (test)

4.16MOS

4

Jun 16, 2026

SEED-TTS en | zh | avg.

0.792SIM (en)

2

Jun 1, 2026

Popular datasets

VCTK Internal LJSpeech Seed-TTS 10-second speech segments

Follow for updates

Join SOTA2 Contact