Views
No views yet
| Metric | Score |
|---|---|
| Accuracy | 0.38 |
| Precision | 0.3075 |
| Recall | 0.38 |
| F1-Score | 0.2871 |
Note: This model is a first iteration and may benefit from further fine-tuning and data augmentation.
0 - angry
1 - apologetic
2 - base
3 - calm
4 - excited
5 - fear
6 - happy
7 - sad
8 - surpriseaudio_path and emotion columns.| Step | Training Loss | Validation Loss |
|---|---|---|
| 0 | 2.23 | 1.91 |
| 350 | 0.88 | 0.86 |
| 525 | 0.98 | 0.53 |
| 750 | 0.22 | 0.35 |
| 1125 | 0.25 | 0.30 |
Full logs available in training script output.
1from transformers import AutoFeatureExtractor, AutoModelForAudioClassification
2import torchaudio
3
4model = AutoModelForAudioClassification.from_pretrained("aicinema69/audio-emotion-detector-v1.0")
5feature_extractor = AutoFeatureExtractor.from_pretrained("aicinema69/audio-emotion-detector-v1.0")
6
7# Load your audio (16kHz recommended)
8waveform, sample_rate = torchaudio.load("your_audio.wav")
9
10# Preprocess
11inputs = feature_extractor(waveform.squeeze().numpy(), sampling_rate=sample_rate, return_tensors="pt", padding=True)
12
13# Predict
14with torch.no_grad():
15 logits = model(**inputs).logits
16 predicted_class = logits.argmax(-1).item()
17
18print("Predicted emotion:", model.config.id2label[predicted_class])trainer.tokenizer (note: tokenizer is deprecated in future 🤗 releases, use processing_class).push_to_hub.1model.push_to_hub("aicinema69/audio-emotion-detector-v1.0")
2feature_extractor.push_to_hub("aicinema69/audio-emotion-detector-v1.0")