metadata
license: mit
base_model: pyannote/segmentation-3.0
tags:
- speaker-diarization
- speaker-segmentation
- generated_from_trainer
datasets:
- diarizers-community/callhome
model-index:
- name: speaker-segmentation-fine-tuned-callhome-jpn
results: []
speaker-segmentation-fine-tuned-callhome-jpn
This model is a fine-tuned version of pyannote/segmentation-3.0 on the diarizers-community/callhome jpn dataset. It achieves the following results on the evaluation set:
- Loss: 0.7579
- Der: 0.2246
- False Alarm: 0.0481
- Missed Detection: 0.1329
- Confusion: 0.0437
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 |
|---|---|---|---|---|---|---|---|
| 0.5686 | 1.0 | 328 | 0.7818 | 0.2346 | 0.0479 | 0.1378 | 0.0489 |
| 0.5307 | 2.0 | 656 | 0.7629 | 0.2278 | 0.0480 | 0.1359 | 0.0440 |
| 0.5212 | 3.0 | 984 | 0.7597 | 0.2287 | 0.0512 | 0.1341 | 0.0435 |
| 0.5155 | 4.0 | 1312 | 0.7562 | 0.2244 | 0.0502 | 0.1314 | 0.0427 |
| 0.5053 | 5.0 | 1640 | 0.7579 | 0.2246 | 0.0481 | 0.1329 | 0.0437 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1