Hugging Face's logo

Tatsapol
/
outs

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
outs / README.md
Tatsapol's picture
Tatsapol
./outs
b5db84f verified
preview code
|
raw
history blame
1.54 kB
metadata
library_name: transformers
license: mit
base_model: biodatlab/distill-whisper-th-small
tags:
  - generated_from_trainer
model-index:
  - name: outs
    results: []

outs

This model is a fine-tuned version of biodatlab/distill-whisper-th-small on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5794
  • Cer: 22.7354

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: 8e-05
  • train_batch_size: 4
  • eval_batch_size: 2
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 6.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Cer
0.3046 1.0 537 0.5714 37.0843
0.1192 2.0 1074 0.5905 29.1700
0.0471 3.0 1611 0.5794 22.7354

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.4.1+cu121
  • Datasets 3.0.1
  • Tokenizers 0.19.1