You can deploy this model and start earning money today!

lucius-40/
speaker-segmentation-fine-tuned-bn

Name: speaker-segmentation-fine-tuned-bn
Availability: InStock

generated_from_trainer

bengali-speaker-diarization

pyannote/segmentation-3.0

finetune

mit

endpoints_compatible

Views

No views yet

speaker-segmentation-fine-tuned-bn-v2

This model is a fine-tuned version of pyannote/segmentation-3.0 on the bengali-speaker-diarization dataset. It achieves the following results on the evaluation set:

Loss: 0.4496
Model Preparation Time: 0.0034
Der: 0.1487
False Alarm: 0.0385
Missed Detection: 0.0281
Confusion: 0.0821

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Der	False Alarm	Missed Detection	Confusion
0.4078	1.0	153	0.4535	0.0034	0.1507	0.0424	0.0168	0.0915
0.3573	2.0	306	0.4361	0.0034	0.1501	0.0376	0.0302	0.0822
0.3413	3.0	459	0.4562	0.0034	0.1495	0.0405	0.0233	0.0857
0.3364	4.0	612	0.4496	0.0034	0.1487	0.0385	0.0281	0.0821

Framework versions

Transformers 4.48.3
Pytorch 2.9.0+cu126
Datasets 3.6.0
Tokenizers 0.21.4

lucius-40/speaker-segmentation-fine-tuned-bn

speaker-segmentation-fine-tuned-bn-v2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

lucius-40/
speaker-segmentation-fine-tuned-bn