# Audio Alignment Detection on AnyAudio-Judge Chinese Subset Source: [Audio Alignment Detection on AnyAudio-Judge Chinese Subset]() Collection index: [llms.txt]() Data updated: 2026-06-04T08:08:32.036Z Evaluation of Audio Alignment Detection performance on the Chinese Subset of the AnyAudio-Judge dataset, assessing quality across Speech, Sound, and Music categories for both real and generated samples. 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 | Audio Alignment Detection | | Dataset (including any stated split) | AnyAudio-Judge Chinese Subset | | Metric | Accuracy (Speech-Real) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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 - [Accuracy (Speech-Real)](): Higher is better; unit: PERCENT. - [Accuracy (Speech-Generated)](): Higher is better; unit: PERCENT. - [Accuracy (Sound-Real)](): Higher is better; unit: PERCENT. - [Accuracy (Sound-Generated)](): Higher is better; unit: PERCENT. - [Accuracy (Music-Real)](): Higher is better; unit: PERCENT. - [Accuracy (Music-Generated)](): Higher is better; unit: PERCENT. - [Accuracy (Mixed)](): Higher is better; unit: PERCENT. - [Average Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Speech-Real) | Result date | Report paper | | --- | --- | --- | --- | | AnyAudio-Judge | 93.33 | Not recorded | 6a21065ac10484bac23efe71 | | Gemini-2.5-Pro | 92.33 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen3-Omni-30B-A3B-Instruct | 90.75 | Not recorded | 6a21065ac10484bac23efe71 | | Gemini-2.5-Pro | 90.67 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen3-Omni-30B-A3B-Captioner | 90.17 | Not recorded | 6a21065ac10484bac23efe71 | | Kimi-Audio-7B-Instruct | 81.17 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen2.5-Omni-7B | 78.17 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen3-Omni-30B-A3B-Captioner | 75.42 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen3-Omni-30B-A3B-Instruct | 67.33 | Not recorded | 6a21065ac10484bac23efe71 | | MiDashengLM | 66.94 | Not recorded | 6a21065ac10484bac23efe71 | | Audio-Flamingo3 | 60.25 | Not recorded | 6a21065ac10484bac23efe71 | | MiDashengLM | 51.58 | Not recorded | 6a21065ac10484bac23efe71 | | Kimi-Audio-7B-Instruct | 51 | Not recorded | 6a21065ac10484bac23efe71 | | Audio-Flamingo3 | 50.25 | Not recorded | 6a21065ac10484bac23efe71 | | Qwen2.5-Omni-7B | 50.25 | Not recorded | 6a21065ac10484bac23efe71 | ### AnyAudio-Judge Accuracy (Speech-Real): 93.33. Result ID: 6a213097fbc373592c1c90da. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Accuracy (Speech-Real): 92.33. Result ID: 6a213097fbc373592c1c90d8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### Qwen3-Omni-30B-A3B-Instruct Accuracy (Speech-Real): 90.75. Result ID: 6a213097fbc373592c1c90d5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Gemini-2.5-Pro Accuracy (Speech-Real): 90.67. Result ID: 6a213097fbc373592c1c90d9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Qwen3-Omni-30B-A3B-Captioner Accuracy (Speech-Real): 90.17. Result ID: 6a213097fbc373592c1c90d7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Kimi-Audio-7B-Instruct Accuracy (Speech-Real): 81.17. Result ID: 6a213097fbc373592c1c90d1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Qwen2.5-Omni-7B Accuracy (Speech-Real): 78.17. Result ID: 6a213097fbc373592c1c90d3. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Qwen3-Omni-30B-A3B-Captioner Accuracy (Speech-Real): 75.42. Result ID: 6a213097fbc373592c1c90d6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### Qwen3-Omni-30B-A3B-Instruct Accuracy (Speech-Real): 67.33. Result ID: 6a213097fbc373592c1c90d4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### MiDashengLM Accuracy (Speech-Real): 66.94. Result ID: 6a213097fbc373592c1c90cf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### Audio-Flamingo3 Accuracy (Speech-Real): 60.25. Result ID: 6a213097fbc373592c1c90cd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"dynamic rubric"}. This record does not assert independent reproduction. ### MiDashengLM Accuracy (Speech-Real): 51.58. Result ID: 6a213097fbc373592c1c90ce. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### Kimi-Audio-7B-Instruct Accuracy (Speech-Real): 51. Result ID: 6a213097fbc373592c1c90d0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### Audio-Flamingo3 Accuracy (Speech-Real): 50.25. Result ID: 6a213097fbc373592c1c90cc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. ### Qwen2.5-Omni-7B Accuracy (Speech-Real): 50.25. Result ID: 6a213097fbc373592c1c90d2. [Reporting paper]() Recorded evaluation configuration: {"Prompt":"holistic"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.