metadata
license: mit
base_model: pyannote/segmentation-3.0
tags:
- speaker-diarization
- speaker-segmentation
- generated_from_trainer
datasets:
- ArtFair/diarizers_dataset_70-15-15
model-index:
- name: results
results: []
results
This model is a fine-tuned version of pyannote/segmentation-3.0 on the ArtFair/diarizers_dataset_70-15-15 default dataset. It achieves the following results on the evaluation set:
- Loss: 0.3941
- Der: 0.2887
- False Alarm: 0.1590
- Missed Detection: 0.1025
- Confusion: 0.0272
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: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Der | False Alarm | Missed Detection | Confusion |
|---|---|---|---|---|---|---|---|
| 0.4009 | 1.0 | 747 | 0.4264 | 0.3191 | 0.1921 | 0.0979 | 0.0291 |
| 0.3626 | 2.0 | 1494 | 0.4017 | 0.3027 | 0.1664 | 0.1085 | 0.0278 |
| 0.3527 | 3.0 | 2241 | 0.4077 | 0.2972 | 0.1354 | 0.1347 | 0.0271 |
| 0.3303 | 4.0 | 2988 | 0.3933 | 0.2867 | 0.1506 | 0.1083 | 0.0277 |
| 0.3312 | 5.0 | 3735 | 0.3941 | 0.2887 | 0.1590 | 0.1025 | 0.0272 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.4.1+cu121
- Datasets 2.17.0
- Tokenizers 0.15.2