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 research benchmarks – Page 5 · SOTA2 Research

Text-to-Speech

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

CV3-Eval Monolingual (hard-en)

4.37WER

7

Jun 8, 2026

CV3-Eval Monolingual (hard-zh)

8.55WER (%)

7

Jun 8, 2026

CV3-Eval Monolingual zh

2.65WER

7

Jun 8, 2026

Seed-TTS Chinese (test)

0.89ZS CER

7

Jun 1, 2026

Seed-TTS hard (test)

7.53WER

7

May 29, 2026

Raon-OpenTTS-1B

6.15WER

7

May 21, 2026

LibriSpeech (test-clean)

0.26Log F0 RMSE (Avg)

7

Apr 2, 2026

1.2WER

7

Apr 2, 2026

LibriTTS-R (test-clean)

1.95CER

7

Mar 10, 2026

2.12WER

7

Jul 2, 2026

Internal Chinese speech dataset 1.0 (test)

4.03NMOS

7

Feb 26, 2026

Text-to-Speech (TTS) Benchmark

1.9WER

7

Feb 26, 2026

Seed-TTS-Eval English (test)

1.83WER (prompting)

6

Jul 7, 2026

MiniMax multilingual zh, en, ja, ko (test)

75TPS

6

Jun 9, 2026

2.2Word Error Rate (WER)

6

May 8, 2026

Headline Set Telugu

0.029LLM-WER

6

Apr 29, 2026

4.45Happy MOS

6

Apr 6, 2026

Chinese and English (test)

0.44CMOS

6

Apr 2, 2026

3.37WER

6

Mar 17, 2026

TextrolSpeech + EmoVoice-DB (test)

4.108MOS-N

6

Mar 17, 2026

VccmDataset (test A)

0.954Pitch

6

Feb 26, 2026

1.9WER

6

Feb 26, 2026

Seed-TTS-Eval hard (test)

6.83WER

6

Feb 26, 2026

10.3WER

6

Feb 26, 2026

0.062Word Error Rate

6

Feb 26, 2026

Page 5 of 13

Popular datasets

LibriSpeech Seed-TTS Multilingual LEMAS-TTS IndicTTS Database

Follow for updates

Join SOTA2 Contact