Whisper Tiny Thai Punctuation 5k - Chee Li

This model is a fine-tuned version of openai/whisper-tiny on the Google Fleurs dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8643
  • Wer: 113.9159

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: 1e-05
  • train_batch_size: 16
  • eval_batch_size: 8
  • 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: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 7000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.2866 5.2356 1000 0.6085 126.8345
0.0843 10.4712 2000 0.6126 116.8844
0.0169 15.7068 3000 0.6997 126.3833
0.0041 20.9424 4000 0.7786 120.2090
0.0019 26.1780 5000 0.8240 116.0294
0.0012 31.4136 6000 0.8532 118.7129
0.0011 36.6492 7000 0.8643 113.9159

Framework versions

  • Transformers 4.46.2
  • Pytorch 2.3.1+cu121
  • Datasets 2.20.0
  • Tokenizers 0.20.3
Downloads last month
16
Safetensors
Model size
37.8M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for CheeLi03/whisper-tiny-th-puct-7k

Finetuned
(1256)
this model

Evaluation results