README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: openai/whisper-large-v2
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: rishabhjain16/infer_cmu_9h
          type: rishabhjain16/infer_cmu_9h
          config: en
          split: test
        metrics:
          - type: wer
            value: 15.22
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: rishabhjain16/infer_pfs
          type: rishabhjain16/infer_pfs
          config: en
          split: test
        metrics:
          - type: wer
            value: 2.88
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: rishabhjain16/infer_myst
          type: rishabhjain16/infer_myst
          config: en
          split: test
        metrics:
          - type: wer
            value: 15.79
            name: WER
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: rishabhjain16/libritts_dev_clean
          type: rishabhjain16/libritts_dev_clean
          config: en
          split: test
        metrics:
          - type: wer
            value: 5.1
            name: WER

openai/whisper-large-v2

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1534
• Wer: 145.6786

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: 32
• 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
0.0799	2.03	500	0.1010	28.1322
0.0239	5.01	1000	0.1388	161.0139
0.0066	7.03	1500	0.1221	99.3747
0.0007	10.01	2000	0.1295	250.8822
0.0007	12.04	2500	0.1423	77.2203
0.0003	15.02	3000	0.1480	149.4380
0.0001	17.05	3500	0.1518	141.5842
0.0001	20.02	4000	0.1534	145.6786

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.1+cu117
• Datasets 2.9.1.dev0
• Tokenizers 0.13.2