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 4 · SOTA2 Research

Text-to-Speech

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

ZTTS1-Eval In-The-Wild (hi)

7.13WER

9

Jun 24, 2026

Microsoft Azure

36.5WER

9

May 28, 2026

Emilia EN speaking-rate

68.8MUSHRA Score

9

Mar 17, 2026

Common Voice en 15

Phi-4 Multimodal

10.8WER

9

Feb 26, 2026

VoxPopuli en V1.0

10.1WER (%)

9

Feb 26, 2026

ZTTS1-Eval In-The-Wild (tl)

15.7WER

8

Jun 24, 2026

ZTTS1-Eval In-The-Wild (ar)

Gemini 3.1 Flash

10.51WER

8

Jun 24, 2026

Baseline (oracle)

1.43WER

8

Jul 2, 2026

SASPEECH (100 samples)

29.4WER

8

Jun 18, 2026

Downstream Audio Generation TTS

3.03WER

8

May 28, 2026

Raon-OpenTTS-Eval (Noisy)

3.42WER

8

May 21, 2026

Raon-OpenTTS Clean (Eval)

Raon-OpenTTS-1B

1.44WER

8

May 21, 2026

Raon-OpenTTS-Eval (Overall)

Raon-OpenTTS-1B

0CMOS

8

May 21, 2026

Raon-OpenTTS-Eval Expressive

0.25CMOS

8

May 21, 2026

Raon-OpenTTS-Eval (Wild)

Raon-OpenTTS-0.3B

0.16CMOS

8

May 21, 2026

Raon-OpenTTS Noisy (eval)

0.31CMOS Score

8

May 21, 2026

Raon-OpenTTS-Eval (Clean)

Raon-OpenTTS-0.3B

0.54CMOS

8

May 21, 2026

Chinese TTS Evaluation ZH (test)

77.3SIM-o

8

Feb 26, 2026

English TTS Evaluation (EN) (test)

0.734SIM-o

8

Feb 26, 2026

Harvard sentences

OverFlow (OFZT)

2.3WER

8

Feb 26, 2026

4.4MOS

8

Feb 26, 2026

ZTTS1-Eval In-The-Wild (th)

3.84WER

7

Jun 24, 2026

ZTTS1 Eval In-The-Wild (id)

5.98WER

7

Jun 24, 2026

SeedTTS en (eval)

1.1WER

7

Jun 23, 2026

NüshuVoice (test)

3.34Naturalness MOS

7

Jun 9, 2026

Page 4 of 13

Popular datasets

LibriSpeech Seed-TTS Multilingual LEMAS-TTS IndicTTS Database

Follow for updates

Join SOTA2 Contact