|
--- |
|
license: apache-2.0 |
|
base_model: facebook/wav2vec2-large-xlsr-53 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- precision |
|
- recall |
|
- f1 |
|
model-index: |
|
- name: SER_wav2vec2-large-xlsr-53_fine-tuned_1.0 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# SER_wav2vec2-large-xlsr-53_240303 |
|
|
|
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on a [Speech Emotion Recognition (en)](https://www.kaggle.com/datasets/dmitrybabko/speech-emotion-recognition-en) dataset. |
|
|
|
This dataset includes the 4 most popular datasets in English: Crema, Ravdess, Savee, and Tess, containing a total of over 12,000 .wav audio files. Each of these four datasets includes 6 to 8 different emotional labels. |
|
|
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.7923 |
|
- Accuracy: 0.2408 |
|
- Precision: 0.2324 |
|
- Recall: 0.2466 |
|
- F1: 0.2226 |
|
|
|
|
|
## For a better performance version, please refer to |
|
|
|
[hughlan1214/Speech_Emotion_Recognition_wav2vec2-large-xlsr-53_240304_SER_fine-tuned2.0](https://huggingface.co/hughlan1214/Speech_Emotion_Recognition_wav2vec2-large-xlsr-53_240304_SER_fine-tuned2.0) |
|
|
|
|
|
## Model description |
|
|
|
The model was obtained through feature extraction using [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) and underwent several rounds of fine-tuning. It predicts the 7 types of emotions contained in speech, aiming to lay the foundation for subsequent use of human micro-expressions on the visual level and context semantics under LLMS to infer user emotions in real-time. |
|
|
|
Although the model was trained on purely English datasets, post-release testing showed that it also performs well in predicting emotions in Chinese and French, demonstrating the powerful cross-linguistic capability of the [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) pre-trained model. |
|
|
|
```python |
|
emotions = ['angry', 'disgust', 'fear', 'happy', 'neutral', 'sad', 'surprise'] |
|
``` |
|
|
|
## Intended uses & limitations |
|
|
|
More information needed |
|
|
|
## Training and evaluation data |
|
|
|
More information needed |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 0.0001 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 3 |
|
- total_train_batch_size: 12 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| |
|
| 1.9297 | 1.0 | 101 | 1.9452 | 0.1233 | 0.0306 | 0.1468 | 0.0454 | |
|
| 1.9114 | 2.0 | 202 | 1.9115 | 0.1773 | 0.1501 | 0.1803 | 0.1323 | |
|
| 1.7863 | 3.0 | 303 | 1.8564 | 0.2081 | 0.1117 | 0.2193 | 0.1336 | |
|
| 1.8439 | 4.0 | 404 | 1.8590 | 0.2042 | 0.2196 | 0.2156 | 0.1755 | |
|
| 1.9361 | 5.0 | 505 | 1.8375 | 0.2081 | 0.2617 | 0.2213 | 0.1573 | |
|
| 1.7572 | 6.0 | 606 | 1.8081 | 0.2100 | 0.2018 | 0.2214 | 0.1841 | |
|
| 1.6715 | 7.0 | 707 | 1.8131 | 0.2389 | 0.2263 | 0.2442 | 0.2129 | |
|
| 1.6687 | 8.0 | 808 | 1.7923 | 0.2408 | 0.2324 | 0.2466 | 0.2226 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.1 |
|
- Pytorch 2.2.1 |
|
- Datasets 2.17.1 |
|
- Tokenizers 0.15.2 |
|
|
