Edit model card

speaker-segmentation-fine-tuned-callhome-spa

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

  • Loss: 0.3513
  • Der: 0.2029
  • False Alarm: 0.1480
  • Missed Detection: 0.0549
  • Confusion: 0.0000

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: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Der False Alarm Missed Detection Confusion
0.3117 1.0 281 0.3448 0.2096 0.1526 0.0545 0.0024
0.2973 2.0 562 0.3260 0.1961 0.1359 0.0601 0.0001
0.2937 3.0 843 0.3413 0.2027 0.1468 0.0555 0.0004
0.2953 4.0 1124 0.3466 0.2023 0.1467 0.0555 0.0000
0.2725 5.0 1405 0.3513 0.2029 0.1480 0.0549 0.0000

Framework versions

  • Transformers 4.45.1
  • Pytorch 2.4.1
  • Datasets 3.0.1
  • Tokenizers 0.20.0
Downloads last month
0
Safetensors
Model size
1.47M params
Tensor type
F32
·
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for mirari/callhomeCompleto

Dataset used to train mirari/callhomeCompleto

Collection including mirari/callhomeCompleto