Audio Classification
Transformers
Safetensors
Slovenian
Croatian
Serbian
wav2vec2-bert
audio-frame-classification
Instructions to use classla/Wav2Vec2BertPrimaryStressAudioFrameClassifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use classla/Wav2Vec2BertPrimaryStressAudioFrameClassifier with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("audio-classification", model="classla/Wav2Vec2BertPrimaryStressAudioFrameClassifier")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForAudioFrameClassification processor = AutoProcessor.from_pretrained("classla/Wav2Vec2BertPrimaryStressAudioFrameClassifier") model = AutoModelForAudioFrameClassification.from_pretrained("classla/Wav2Vec2BertPrimaryStressAudioFrameClassifier", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -45,8 +45,13 @@ This model annotates primary stress in words on 20ms frames.
|
|
| 45 |
note = {Accepted at Interspeech 2025}
|
| 46 |
}
|
| 47 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
### Evaluation results
|
| 49 |
|
|
|
|
| 50 |
|
| 51 |
|test language|accuracy|
|
| 52 |
| ---|---|
|
|
|
|
| 45 |
note = {Accepted at Interspeech 2025}
|
| 46 |
}
|
| 47 |
```
|
| 48 |
+
### Training data
|
| 49 |
+
|
| 50 |
+
The model was trained on the training split of [ParlaStress-HR dataset](http://hdl.handle.net/11356/2038).
|
| 51 |
+
|
| 52 |
### Evaluation results
|
| 53 |
|
| 54 |
+
For evaluation, the test splits of [ParlaStress-HR dataset](http://hdl.handle.net/11356/2038) were used.
|
| 55 |
|
| 56 |
|test language|accuracy|
|
| 57 |
| ---|---|
|