Zero-shot Text-to-Speech
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
1.08WER
25
Jun 2, 2026
4.9WER
15
Feb 26, 2026
5.1WER
15
Feb 26, 2026
0.038WER
15
Feb 26, 2026
0.764SECS (s)
12
Mar 5, 2026
0.765SECS
12
Mar 5, 2026
1.78WER
12
Feb 26, 2026
3.89UTMOS
10
Jun 17, 2026
1.6WER (Whisper-large)
8
May 29, 2026
0.84WER
8
May 8, 2026
7.66WER
8
Mar 6, 2026
10.91WER
8
Mar 6, 2026
4WER
8
Feb 26, 2026
3.5WER
8
Feb 26, 2026
4.7WER
8
Feb 26, 2026
1.65WER
7
Mar 17, 2026
1.3WER
6
Feb 26, 2026
2.58UTMOS
6
Feb 26, 2026
2.9UTMOS
6
Feb 26, 2026
3.09UTMOS
6
Feb 26, 2026
3.76UTMOS
6
Feb 26, 2026
4.87English CER/WER
5
May 26, 2026
4.31MOS (Naturalness)
5
Mar 5, 2026
1.9WER
5
Feb 26, 2026
4.64IMOS
4
May 26, 2026