Update README.md
Browse files
README.md
CHANGED
|
@@ -1,41 +1,43 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
datasets:
|
| 4 |
-
- stapesai/ssi-speech-emotion-recognition
|
| 5 |
-
metrics:
|
| 6 |
-
- accuracy
|
| 7 |
-
- precision
|
| 8 |
-
- recall
|
| 9 |
-
- f1
|
| 10 |
-
pipeline_tag: audio-classification
|
| 11 |
-
|
| 12 |
-
|
| 13 |
-
|
| 14 |
-
|
| 15 |
-
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
- **
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
|
| 39 |
-
|
| 40 |
-
|
|
|
|
|
|
|
| 41 |
**Overall Accuracy**: 87%
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
datasets:
|
| 4 |
+
- stapesai/ssi-speech-emotion-recognition
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
- precision
|
| 8 |
+
- recall
|
| 9 |
+
- f1
|
| 10 |
+
pipeline_tag: audio-classification
|
| 11 |
+
base_model:
|
| 12 |
+
- facebook/wav2vec2-base
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# Multimodal Emotion Speech Recognition
|
| 16 |
+
|
| 17 |
+
## Model Description
|
| 18 |
+
This model performs emotion recognition from speech using a multimodal approach, utilizing:
|
| 19 |
+
- **Audio Model**: Wav2Vec2 Base
|
| 20 |
+
## Dataset
|
| 21 |
+
- **Dataset Name**: [stapesai/ssi-speech-emotion-recognition](https://huggingface.co/datasets/stapesai/ssi-speech-emotion-recognition)
|
| 22 |
+
|
| 23 |
+
## Evaluation Results
|
| 24 |
+
|
| 25 |
+
### Classification Report
|
| 26 |
+
```
|
| 27 |
+
precision recall f1-score support
|
| 28 |
+
|
| 29 |
+
ANG 0.97 0.93 0.95 30
|
| 30 |
+
CAL 0.00 0.00 0.00 0
|
| 31 |
+
DIS 0.95 0.90 0.92 20
|
| 32 |
+
FEA 0.76 0.70 0.73 27
|
| 33 |
+
HAP 0.87 0.82 0.84 33
|
| 34 |
+
NEU 0.96 0.96 0.96 25
|
| 35 |
+
SAD 0.73 1.00 0.84 19
|
| 36 |
+
SUR 0.88 0.78 0.82 9
|
| 37 |
+
|
| 38 |
+
accuracy 0.87 163
|
| 39 |
+
macro avg 0.76 0.76 0.76 163
|
| 40 |
+
weighted avg 0.88 0.87 0.87 163
|
| 41 |
+
```
|
| 42 |
+
|
| 43 |
**Overall Accuracy**: 87%
|