Text-to-Speech
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
5.09Character Error Rate (CER)
17
Jun 23, 2026
1.98WER (%)
17
Feb 26, 2026
0.89CER
16
Feb 26, 2026
0DTW
16
Feb 26, 2026
4.54MOS
16
Jun 1, 2026
2.43WER
15
Jun 8, 2026
75Score (Angry)
15
Jun 2, 2026
5.71CER
15
Apr 1, 2026
3.956UTMOS
15
May 25, 2026
92.3APS
15
Feb 26, 2026
4.47MOS
15
Jun 16, 2026
1.39WER
14
Jun 23, 2026
1.47WER
14
Apr 27, 2026
4.303UTMOS
14
Mar 5, 2026
9.9WER
12
Jul 3, 2026
8.5WER (%)
12
Jul 3, 2026
7.5WER
12
Jul 3, 2026
10WER
12
Jul 3, 2026
42.2WER
12
Jul 3, 2026
1.95WER
12
Jun 9, 2026
6.83CER
12
May 19, 2026
95.8Accuracy
12
Apr 28, 2026
94.9Accuracy
12
Apr 28, 2026
1.69WER
12
Feb 26, 2026
1.07WER
12
Apr 15, 2026