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: 0.7498
Der: 0.2258
False Alarm: 0.0470
Missed Detection: 0.1339
Confusion: 0.0449

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
0.5797	1.0	328	0.7543	0.2334	0.0582	0.1277	0.0476
0.5578	2.0	656	0.7684	0.2306	0.0446	0.1401	0.0458
0.5213	3.0	984	0.7569	0.2289	0.0463	0.1368	0.0459
0.4971	4.0	1312	0.7448	0.2273	0.0489	0.1329	0.0456
0.5026	5.0	1640	0.7498	0.2258	0.0470	0.1339	0.0449

Framework versions

Transformers 4.40.0
Pytorch 2.2.2+cu121
Datasets 2.18.0
Tokenizers 0.19.1