README.md

metadata

license: apache-2.0
base_model: openai/whisper-tiny
tags:
  - generated_from_trainer
datasets:
  - common_voice_9_0
metrics:
  - wer
model-index:
  - name: cv9-special-batch8-tiny
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_9_0
          type: common_voice_9_0
          config: id
          split: test
          args: id
        metrics:
          - name: Wer
            type: wer
            value: 31.874856222682308

cv9-special-batch8-tiny

This model is a fine-tuned version of openai/whisper-tiny on the common_voice_9_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4986
• Wer: 31.8749

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: 8
• eval_batch_size: 4
• 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: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.6281	0.97	1000	0.5817	37.6950
0.4018	1.94	2000	0.5157	34.2121
0.2914	2.9	3000	0.4980	32.4960
0.2078	3.87	4000	0.4968	31.7506
0.1925	4.84	5000	0.4986	31.8749

Framework versions

• Transformers 4.31.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3