# Multimodal Emotion Analysis on IEMOCAP (word aligned)

Source: [Multimodal Emotion Analysis on IEMOCAP (word aligned)](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned>)

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

Data updated: 2026-02-26T06:07:33.302Z

Evaluation of Multimodal Emotion Analysis performance using the word-aligned subset of the IEMOCAP dataset, reporting per-emotion accuracy, F1 scores, and associated training/testing losses.

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 | Multimodal emotion analysis |
| Dataset (including any stated split) | IEMOCAP (word aligned) |
| Metric | Train Loss |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 2 of 2 |
| 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

- [Train Loss](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Train%20Loss>): Lower is better; unit: SCALAR.

- [Test Loss](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Test%20Loss>): Lower is better; unit: SCALAR.

- [Happy Accuracy](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Happy%20Accuracy>): Higher is better; unit: PERCENT.

- [Happy F1 Score](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Happy%20F1%20Score>): Higher is better; unit: PERCENT.

- [Sad Accuracy](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Sad%20Accuracy>): Higher is better; unit: PERCENT.

- [Sad F1 Score](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Sad%20F1%20Score>): Higher is better; unit: PERCENT.

- [Angry Accuracy](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Angry%20Accuracy>): Higher is better; unit: PERCENT.

- [Angry F1 Score](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Angry%20F1%20Score>): Higher is better; unit: PERCENT.

- [Natural Accuracy](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Natural%20Accuracy>): Higher is better; unit: PERCENT.

- [Natural F1 Score](<https://www.sota2.com/research/sota/multimodal-emotion-analysis-on-iemocap-word-aligned.md?metric=Natural%20F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Train Loss | Result date | Report paper |
| --- | --- | --- | --- |
| Multimodal Transformer (MulT) + RegBN | 0.009 | Not recorded | 68fa71411e209131cea502c6 |
| Multimodal Transformer (MulT) | 0.106 | Not recorded | 68fa71411e209131cea502c6 |

### Multimodal Transformer (MulT) + RegBN

Train Loss: 0.009. Result ID: 6991c1c93a6c3148490a6093.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71411e209131cea502c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71411e209131cea502c6.md>)

[Implementation](<https://mogvision.github.io/RegBN>)

Recorded evaluation configuration: {"normalization":"RegBN"}. This record does not assert independent reproduction.

### Multimodal Transformer (MulT)

Train Loss: 0.106. Result ID: 6991c1c93a6c3148490a6092.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71411e209131cea502c6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa65d61e209131cea0c91f.md>)

[Implementation](<https://github.com/yaohungt/Multimodal-Transformer>)

Recorded evaluation configuration: {"normalization":"Baseline"}. This record does not assert independent reproduction.

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