README.md

metadata

library_name: transformers
language:
  - th
license: apache-2.0
base_model: biodatlab/whisper-th-medium-combined
tags:
  - asr
  - speech-recognition
  - thai
  - custom-model
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: Whisper Medium TH - Nonhmello
    results: []

Whisper Medium TH - Nonhmello

This model is a fine-tuned version of biodatlab/whisper-th-medium-combined on the Custom dataset on local machine dataset. It achieves the following results on the evaluation set:

• Loss: 0.3068
• Wer: 77.4194

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: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0092	83.3333	500	0.2586	90.3226
0.0006	166.6667	1000	0.3068	77.4194

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3