# Paralinguistics on IEMOCAP (LB) Source: [Paralinguistics on IEMOCAP (LB)]() Collection index: [llms.txt]() Data updated: 2026-05-12T16:03:30.183Z Evaluation of Paralinguistics performance on the IEMOCAP dataset, reporting the LB metric. 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 | Paralinguistics | | Dataset (including any stated split) | IEMOCAP | | Metric | LB | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 10 of 10 | | 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 - [LB](): Higher is better; unit: SCALAR. ## Recorded results | Method | LB | Result date | Report paper | | --- | --- | --- | --- | | Gemini2.5-Flash | 92.7 | Not recorded | 68fa7b4a1e209131cea87950 | | Kimi-Audio | 89 | Not recorded | 68fa7b4a1e209131cea87950 | | Qwen2.5-Omni-7B | 85.8 | Not recorded | 68fa7b4a1e209131cea87950 | | Qwen3-Omni-30B-A3B-Thinking | 82.5 | Not recorded | 68fa7b4a1e209131cea87950 | | Qwen2.5-Omni-3B | 81.5 | Not recorded | 68fa7b4a1e209131cea87950 | | Voxtral-Mini-3B | 54.9 | Not recorded | 68fa7b4a1e209131cea87950 | | Phi-4-Multi-modal | 50.5 | Not recorded | 68fa7b4a1e209131cea87950 | | Voxtral-Small-24B | 42.8 | Not recorded | 68fa7b4a1e209131cea87950 | | GPT-4o-transcribe + GPT-4.1-mini | 30.2 | Not recorded | 68fa7b4a1e209131cea87950 | | Whisper-Large-v3 + GPT-oSS-20B | 1.9 | Not recorded | 68fa7b4a1e209131cea87950 | ### Gemini2.5-Flash LB: 92.7. Result ID: 6a034f3f47f34ce924e8a623. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Proprietary Audio Language Models"}. This record does not assert independent reproduction. ### Kimi-Audio LB: 89. Result ID: 6a034f3f47f34ce924e8a620. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medium Sized Large Audio Language Models","Parameter Count":"5B-20B parameters"}. This record does not assert independent reproduction. ### Qwen2.5-Omni-7B LB: 85.8. Result ID: 6a034f3f47f34ce924e8a61f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Medium Sized Large Audio Language Models","Parameter Count":"5B-20B parameters"}. This record does not assert independent reproduction. ### Qwen3-Omni-30B-A3B-Thinking LB: 82.5. Result ID: 6a034f3f47f34ce924e8a622. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Large Sized Large Audio Language Models","Parameter Count":"\>20B parameters"}. This record does not assert independent reproduction. ### Qwen2.5-Omni-3B LB: 81.5. Result ID: 6a034f3f47f34ce924e8a61d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Small-sized Audio Language Models","Parameter Count":"\<5B parameters"}. This record does not assert independent reproduction. ### Voxtral-Mini-3B LB: 54.9. Result ID: 6a034f3f47f34ce924e8a61c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Small-sized Audio Language Models","Parameter Count":"\<5B parameters"}. This record does not assert independent reproduction. ### Phi-4-Multi-modal LB: 50.5. Result ID: 6a034f3f47f34ce924e8a61e. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Medium Sized Large Audio Language Models","Parameter Count":"5B-20B parameters"}. This record does not assert independent reproduction. ### Voxtral-Small-24B LB: 42.8. Result ID: 6a034f3f47f34ce924e8a621. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Large Sized Large Audio Language Models","Parameter Count":"\>20B parameters"}. This record does not assert independent reproduction. ### GPT-4o-transcribe + GPT-4.1-mini LB: 30.2. Result ID: 6a034f3f47f34ce924e8a625. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Cascaded Systems"}. This record does not assert independent reproduction. ### Whisper-Large-v3 + GPT-oSS-20B LB: 1.9. Result ID: 6a034f3f47f34ce924e8a624. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Cascaded Systems"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.