README.md

metadata

language:
  - ivn
license: apache-2.0
base_model: openai/whisper-large-v3
tags:
  - generated_from_trainer
datasets:
  - audiofolder
metrics:
  - wer
model-index:
  - name: whisper-large-v3-ivn-v1
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: audiofolder
          type: audiofolder
          config: default
          split: train
          args: 'config: ivn, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 70.56790998493842

whisper-large-v3-ivn-v1

This model is a fine-tuned version of openai/whisper-large-v3 on the audiofolder dataset. It achieves the following results on the evaluation set:

• Loss: 1.8302
• Wer: 70.5679

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0274	14.29	1000	1.4520	78.4974
0.0033	28.57	2000	1.6206	73.4296
0.0004	42.86	3000	1.7704	70.3553
0.0002	57.14	4000	1.8302	70.5679

Framework versions

• Transformers 4.37.1
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.1