This model is a fine-tuned version of pyannote/speaker-diarization-3.1 on the diarizers-community/callhome dataset. It achieves the following results on the evaluation set:

Loss: 0.4388
Der: 0.1470
False Alarm: 0.0241
Missed Detection: 0.0294
Confusion: 0.0934

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 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Der	False Alarm	Missed Detection	Confusion
0.4572	1.0	194	0.4811	0.1598	0.0239	0.0319	0.1041
0.3809	2.0	388	0.4470	0.1488	0.0223	0.0315	0.0950
0.3892	3.0	582	0.4388	0.1470	0.0241	0.0294	0.0934

Framework versions

Transformers 4.48.1
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0

abhimehra8194/speaker-segmentation-fine-tuned-callhome-hindi-2

speaker-segmentation-fine-tuned-callhome-hi

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

abhimehra8194/
speaker-segmentation-fine-tuned-callhome-hindi-2