README.md

metadata

library_name: transformers
language:
  - en
license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - generated_from_trainer
datasets:
  - wwwtwwwt/fineaudio-ScienceTechnology
metrics:
  - wer
model-index:
  - name: Whisper Tiny En - ScienceTechnology - AI Concepts
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: fineaudio-ScienceTechnology-AI Concepts
          type: wwwtwwwt/fineaudio-ScienceTechnology
          args: 'config: en, split: test'
        metrics:
          - name: Wer
            type: wer
            value: 32.08077778075368

Whisper Tiny En - ScienceTechnology - AI Concepts

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

• Loss: 0.5842
• Wer: 32.0808

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: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.522	1.2690	1000	0.6057	37.2097
0.3557	2.5381	2000	0.5705	31.0737
0.2384	3.8071	3000	0.5771	31.5585
0.2008	5.0761	4000	0.5842	32.0808

Framework versions

• Transformers 4.46.3
• Pytorch 2.4.0
• Datasets 3.1.0
• Tokenizers 0.20.0