Target-Speaker Multilingual Speech Generation on TTS multilingual Spanish (test)
1.154WERQwen3-TTS
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen3-TTSTokenizer Frequency=12Hz, Parameters=0.6B, Evaluation Protocol=CustomVoice2026.01 | 1.154 | |
| Qwen3-TTSTokenizer Frequency=25Hz, Parameters=1.7B, Evaluation Protocol=CustomVoice2026.01 | 1.284 | |
| Qwen3-TTSTokenizer Frequency=25Hz, Parameters=0.6B, Evaluation Protocol=CustomVoice2026.01 | 1.309 | |
| Qwen3-TTSTokenizer Frequency=12Hz, Parameters=1.7B, Evaluation Protocol=CustomVoice2026.01 | 1.33 | |
| GPT-4o-Audio Preview2026.01 | 4 |