---
license: apache-2.0
base_model: hughlan1214/SER_wav2vec2-large-xlsr-53_240304_fine-tuned1.1
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: hughlan1214/Speech_Emotion_Recognition_wav2vec2-large-xlsr-53_240304_SER_fine-tuned2.0
  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_240304_fine-tuned_2

This model is a fine-tuned version of [hughlan1214/SER_wav2vec2-large-xlsr-53_240304_fine-tuned1.1](https://huggingface.co/hughlan1214/SER_wav2vec2-large-xlsr-53_240304_fine-tuned1.1) 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.

This achieves the following results on the evaluation set:
- Loss: 1.0601
- Accuracy: 0.6731
- Precision: 0.6761
- Recall: 0.6794
- F1: 0.6738

## 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

70/30 of entire dataset.

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Accuracy | Precision | Recall | F1     |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:---------:|:------:|:------:|
| 0.8904        | 1.0   | 1048  | 1.1923          | 0.5773   | 0.6162    | 0.5563 | 0.5494 |
| 1.1394        | 2.0   | 2096  | 1.0143          | 0.6071   | 0.6481    | 0.6189 | 0.6057 |
| 0.9373        | 3.0   | 3144  | 1.0585          | 0.6126   | 0.6296    | 0.6254 | 0.6119 |
| 0.7405        | 4.0   | 4192  | 0.9580          | 0.6514   | 0.6732    | 0.6562 | 0.6576 |
| 1.1638        | 5.0   | 5240  | 0.9940          | 0.6486   | 0.6485    | 0.6627 | 0.6435 |
| 0.6741        | 6.0   | 6288  | 1.0307          | 0.6628   | 0.6710    | 0.6711 | 0.6646 |
| 0.604         | 7.0   | 7336  | 1.0248          | 0.6667   | 0.6678    | 0.6751 | 0.6682 |
| 0.6835        | 8.0   | 8384  | 1.0396          | 0.6722   | 0.6803    | 0.6790 | 0.6743 |
| 0.5421        | 9.0   | 9432  | 1.0493          | 0.6714   | 0.6765    | 0.6785 | 0.6736 |
| 0.5728        | 10.0  | 10480 | 1.0601          | 0.6731   | 0.6761    | 0.6794 | 0.6738 |


### Framework versions

- Transformers 4.38.1
- Pytorch 2.2.1
- Datasets 2.17.1
- Tokenizers 0.15.2