mrm8488/speaker-segmentation-fine-tuned-callhome-spa
This model is a fine-tuned version of openai/whisper-small on the diarizers-community/callhome dataset. It achieves the following results on the evaluation set:
- Loss: 0.5179
- Der: 0.1717
- False Alarm: 0.0717
- Missed Detection: 0.0687
- Confusion: 0.0312
Model description
More information needed
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: 0.001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Der | False Alarm | Missed Detection | Confusion |
|---|---|---|---|---|---|---|---|
| 0.6432 | 1.0 | 382 | 0.5219 | 0.1750 | 0.0646 | 0.0755 | 0.0349 |
| 0.6133 | 2.0 | 764 | 0.5387 | 0.1821 | 0.0749 | 0.0717 | 0.0356 |
| 0.615 | 3.0 | 1146 | 0.5146 | 0.1729 | 0.0748 | 0.0666 | 0.0315 |
| 0.6268 | 4.0 | 1528 | 0.5220 | 0.1727 | 0.0711 | 0.0690 | 0.0326 |
| 0.6037 | 5.0 | 1910 | 0.5179 | 0.1717 | 0.0717 | 0.0687 | 0.0312 |
Framework versions
- Transformers 4.40.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
- Downloads last month
- 8
Model tree for mrm8488/peaker-segmentation-fine-tuned-callhome-spa
Base model
openai/whisper-small