# Spoken Dialogue Evaluation on C3 ZH

Source: [Spoken Dialogue Evaluation on C3 ZH](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh>)

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

Data updated: 2026-05-11T07:28:48.506Z

Evaluation of spoken dialogue performance on the C3 ZH dataset, utilizing metrics covering phonetic, semantic, coreference, and multi-turn conversational aspects.

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 | Spoken Dialogue Evaluation |
| Dataset (including any stated split) | C3 ZH |
| Metric | Phonetic Error |
| Unit | PERCENT |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| 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

- [Phonetic Error](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Phonetic%20Error>): Lower is better; unit: PERCENT.

- [Semantic Error](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Semantic%20Error>): Lower is better; unit: PERCENT.

- [Ambiguity Error](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Ambiguity%20Error>): Lower is better; unit: PERCENT.

- [Omission Error](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Omission%20Error>): Lower is better; unit: PERCENT.

- [Coreference Score](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Coreference%20Score>): Higher is better; unit: PERCENT.

- [Multi-Turn Score](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Multi-Turn%20Score>): Higher is better; unit: PERCENT.

- [Context Dependency Score](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Context%20Dependency%20Score>): Higher is better; unit: PERCENT.

- [Overall Score](<https://www.sota2.com/research/sota/spoken-dialogue-evaluation-on-c3-zh.md?metric=Overall%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Phonetic Error | Result date | Report paper |
| --- | --- | --- | --- |
| VITA-QINYU-4B | 9.19 | Not recorded | 6a016286dbc836b87d82302e |
| VITA-QINYU-8B | 13.51 | Not recorded | 6a016286dbc836b87d82302e |
| Freeze-Omni | 16.22 | Not recorded | 6a016286dbc836b87d82302e |
| Step-Audio | 18.92 | Not recorded | 6a016286dbc836b87d82302e |
| Kimi-Audio | 20.27 | Not recorded | 6a016286dbc836b87d82302e |
| GLM-4-Voice | 22.97 | Not recorded | 6a016286dbc836b87d82302e |
| Qwen2.5-Omni | 27.03 | Not recorded | 6a016286dbc836b87d82302e |

### VITA-QINYU-4B

Phonetic Error: 9.19. Result ID: 6a01846e47f34ce924e2f66f.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

Recorded evaluation configuration: {"#Param":"4B"}. This record does not assert independent reproduction.

### VITA-QINYU-8B

Phonetic Error: 13.51. Result ID: 6a01846e47f34ce924e2f670.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

Recorded evaluation configuration: {"#Param":"8B"}. This record does not assert independent reproduction.

### Freeze-Omni

Phonetic Error: 16.22. Result ID: 6a01846e47f34ce924e2f66e.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa760a1e209131cea6b0d4.md>)

Recorded evaluation configuration: {"#Param":"7B"}. This record does not assert independent reproduction.

### Step-Audio

Phonetic Error: 18.92. Result ID: 6a01846e47f34ce924e2f66a.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa795a1e209131cea7d396.md>)

Recorded evaluation configuration: {"#Param":"130B"}. This record does not assert independent reproduction.

### Kimi-Audio

Phonetic Error: 20.27. Result ID: 6a01846e47f34ce924e2f66c.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78431e209131cea77803.md>)

Recorded evaluation configuration: {"#Param":"7B"}. This record does not assert independent reproduction.

### GLM-4-Voice

Phonetic Error: 22.97. Result ID: 6a01846e47f34ce924e2f66d.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75f91e209131cea6abba.md>)

Recorded evaluation configuration: {"#Param":"9B"}. This record does not assert independent reproduction.

### Qwen2.5-Omni

Phonetic Error: 27.03. Result ID: 6a01846e47f34ce924e2f66b.

[Reporting paper](<https://www.sota2.com/research/paper/6a016286dbc836b87d82302e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa767e1e209131cea6d6a7.md>)

Recorded evaluation configuration: {"#Param":"7B"}. This record does not assert independent reproduction.

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