# Voice Cloning on Seed-TTS-Eval ZH-Hard (test)

Source: [Voice Cloning on Seed-TTS-Eval ZH-Hard (test)](<https://www.sota2.com/research/sota/voice-cloning-on-seed-tts-eval-zh-hard-test>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-08T07:04:44.418Z

Evaluation of Voice Cloning performance on the challenging Chinese subset ('ZH-Hard') test split of the Seed-TTS-Eval dataset, measured using Character Error Rate (CER) and SIM.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Voice Cloning |
| Dataset (including any stated split) | Seed-TTS-Eval ZH-Hard (test) |
| Metric | CER |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 18 of 18 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [CER](<https://www.sota2.com/research/sota/voice-cloning-on-seed-tts-eval-zh-hard-test.md?metric=CER>): Lower is better; unit: PERCENT.

- [SIM](<https://www.sota2.com/research/sota/voice-cloning-on-seed-tts-eval-zh-hard-test.md?metric=SIM>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CER | Result date | Report paper |
| --- | --- | --- | --- |
| CosyVoice3.5 | 5.71 | Not recorded | 6a264c5590966ade919be7d1 |
| CosyVoice 3 | 5.83 | Not recorded | 6a264c5590966ade919be7d1 |
| Fish Audio S2 | 5.99 | Not recorded | 6a264c5590966ade919be7d1 |
| LongCat-Audio-DiT | 6.04 | Not recorded | 6a264c5590966ade919be7d1 |
| CosyVoice 3 | 6.08 | Not recorded | 6a264c5590966ade919be7d1 |
| Qwen3-TTS | 6.76 | Not recorded | 6a264c5590966ade919be7d1 |
| CosyVoice 2 | 6.83 | Not recorded | 6a264c5590966ade919be7d1 |
| IndexTTS2 | 7.12 | Not recorded | 6a264c5590966ade919be7d1 |
| Seed-TTS | 7.59 | Not recorded | 6a264c5590966ade919be7d1 |
| VoxCPM1.5 | 7.74 | Not recorded | 6a264c5590966ade919be7d1 |
| Qwen2.5-Omni | 7.97 | Not recorded | 6a264c5590966ade919be7d1 |
| VoxCPM2 | 8.13 | Not recorded | 6a264c5590966ade919be7d1 |
| F5-TTS | 8.67 | Not recorded | 6a264c5590966ade919be7d1 |
| VoxCPM | 8.87 | Not recorded | 6a264c5590966ade919be7d1 |
| CosyVoice | 11.75 | Not recorded | 6a264c5590966ade919be7d1 |
| FireRedTTS | 17.45 | Not recorded | 6a264c5590966ade919be7d1 |
| OpenAudio-s1-mini | 23.37 | Not recorded | 6a264c5590966ade919be7d1 |
| HiggsAudio-v2 | 55.07 | Not recorded | 6a264c5590966ade919be7d1 |

### CosyVoice3.5

CER: 5.71. Result ID: 6a266989fbc373592c22a975.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

Recorded evaluation configuration: {"Open-source":false,"Zero-shot":true}. This record does not assert independent reproduction.

### CosyVoice 3

CER: 5.83. Result ID: 6a266989fbc373592c22a973.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78dd1e209131cea7ae5b.md>)

Recorded evaluation configuration: {"Parameters":"1.5B","Open-source":false,"Zero-shot":true}. This record does not assert independent reproduction.

### Fish Audio S2

CER: 5.99. Result ID: 6a266989fbc373592c22a980.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b0f6e92d603e6de3e45b93.md>)

[Implementation](<https://github.com/fishaudio/fish-speech>)

Recorded evaluation configuration: {"Parameters":"4B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### LongCat-Audio-DiT

CER: 6.04. Result ID: 6a266989fbc373592c22a981.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69cca6bf0636e5bf4afcbd4f.md>)

Recorded evaluation configuration: {"Parameters":"3.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### CosyVoice 3

CER: 6.08. Result ID: 6a266989fbc373592c22a979.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78dd1e209131cea7ae5b.md>)

Recorded evaluation configuration: {"Parameters":"0.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### Qwen3-TTS

CER: 6.76. Result ID: 6a266989fbc373592c22a97f.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6973006ab2be0ed85a78c8c5.md>)

[Implementation](<https://github.com/QwenLM/Qwen3-TTS>)

Recorded evaluation configuration: {"Parameters":"1.7B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### CosyVoice 2

CER: 6.83. Result ID: 6a266989fbc373592c22a978.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76591e209131cea6c851.md>)

Recorded evaluation configuration: {"Parameters":"0.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### IndexTTS2

CER: 7.12. Result ID: 6a266989fbc373592c22a97d.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b1a1e209131cea86926.md>)

Recorded evaluation configuration: {"Parameters":"1.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### Seed-TTS

CER: 7.59. Result ID: 6a266989fbc373592c22a974.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa736f1e209131cea5c769.md>)

Recorded evaluation configuration: {"Open-source":false,"Zero-shot":true}. This record does not assert independent reproduction.

### VoxCPM1.5

CER: 7.74. Result ID: 6a266989fbc373592c22a983.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

Recorded evaluation configuration: {"Parameters":"0.8B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### Qwen2.5-Omni

CER: 7.97. Result ID: 6a266989fbc373592c22a97b.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77a21e209131cea7482c.md>)

Recorded evaluation configuration: {"Parameters":"7B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### VoxCPM2

CER: 8.13. Result ID: 6a266989fbc373592c22a984.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

Recorded evaluation configuration: {"Parameters":"2B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### F5-TTS

CER: 8.67. Result ID: 6a266989fbc373592c22a976.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78a71e209131cea79a4f.md>)

[Implementation](<https://github.com/SWivid/F5-TTS>)

Recorded evaluation configuration: {"Parameters":"0.3B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### VoxCPM

CER: 8.87. Result ID: 6a266989fbc373592c22a982.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7bb81e209131cea8a230.md>)

Recorded evaluation configuration: {"Parameters":"0.6B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### CosyVoice

CER: 11.75. Result ID: 6a266989fbc373592c22a977.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74031e209131cea5f3e0.md>)

Recorded evaluation configuration: {"Parameters":"0.3B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### FireRedTTS

CER: 17.45. Result ID: 6a266989fbc373592c22a97a.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77f31e209131cea762eb.md>)

Recorded evaluation configuration: {"Parameters":"0.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### OpenAudio-s1-mini

CER: 23.37. Result ID: 6a266989fbc373592c22a97c.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

Recorded evaluation configuration: {"Parameters":"0.5B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

### HiggsAudio-v2

CER: 55.07. Result ID: 6a266989fbc373592c22a97e.

[Reporting paper](<https://www.sota2.com/research/paper/6a264c5590966ade919be7d1.md>)

Recorded evaluation configuration: {"Parameters":"3B","Open-source":true,"Zero-shot":true}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
