The Interactive Emotional Dyadic Motion Capture (IEMOCAP) Database
Multimodal Emotion Recognition IEMOCAP The IEMOCAP dataset consists of 151 videos of recorded dialogues, with 2 speakers per session for a total of 302 videos across the dataset. Each segment is annotated for the presence of 9 emotions (angry, excited, fear, sad, surprised, frustrated, happy, disappointed and neutral) as well as valence, arousal and dominance. The dataset is recorded across 5 sessions with 5 pairs of speakers.
Source: Multi-attention Recurrent Network for Human Communication Comprehension
Image Source: https://sail.usc.edu/iemocap/Busso_2008_iemocap.pdf
Variants: IEMOCAP
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Multimodal Emotion Recognition | GraphSmile | Tracing Intricate Cues in Dialogue: … | 2024-07-31 |
| Speech Emotion Recognition | emoDARTS | emoDARTS: Joint Optimisation of CNN … | 2024-03-21 |
| Multimodal Emotion Recognition | Joyful | Joyful: Joint Modality Fusion and … | 2023-11-18 |
| Speech Emotion Recognition | CNN - DARTS | Enhancing Speech Emotion Recognition Through … | 2023-05-23 |
| Speech Emotion Recognition | TAP | Speaker Normalization for Self-supervised Speech … | 2022-02-02 |
| Speech Emotion Recognition | Partially Fine-tuned HuBERT Large | A Fine-tuned Wav2vec 2.0/HuBERT Benchmark … | 2021-11-04 |
| Speech Emotion Recognition | CNN+LSTM | CNN+LSTM Architecture for Speech Emotion … | 2018-02-15 |
Recent papers with results on this dataset: