Minimax
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Text-to-Speech | MiniMax Multilingual 24 (test) | WER0.572 | 75 | |
| Persona Discrimination | MiniMax Cross-generator M2.5 | Persona Separability (Δ)0.281 | 16 | |
| Speech Synthesis | Minimax Multilingual (test) | Chinese WER (%)0.73 | 9 | |
| Speech Generation | MiniMax ko | CER1.57 | 7 | |
| Text-to-Speech | MiniMax multilingual zh, en, ja, ko (test) | TPS75 | 6 | |
| Speech Synthesis | MiniMax multilingual Korean (test) | Word Error Rate1.75 | 4 | |
| Speech Synthesis | MiniMax multilingual Japanese (test) | Word Error Rate (WER)3.52 | 4 | |
| Speech Synthesis | MiniMax multilingual English (test) | Word Error Rate2.16 | 4 | |
| Speech Synthesis | MiniMax multilingual Chinese (test) | WER1.08 | 4 | |
| Speech Synthesis | MiniMax multilingual German (test) | WER0.57 | 3 | |
| Speech Synthesis | MiniMax multilingual French (test) | WER4.1 | 3 |
Showing 11 of 11 rows