metadata

language:
  - en
license: apache-2.0
base_model: openai/whisper-tiny.en
tags:
  - generated_from_trainer
datasets:
  - Dev372/Medical_STT_Dataset_1.1
metrics:
  - wer
model-index:
  - name: English Whisper Model
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: Medical
          type: Dev372/Medical_STT_Dataset_1.1
          args: 'split: test'
        metrics:
          - name: Wer
            type: wer
            value: 6.5482216924132075

English Whisper Model

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

Loss: 0.1566
Wer: 6.5482

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: 18
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: 1100
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.8857	0.1554	55	1.6694	13.1520
1.3264	0.3107	110	1.0577	11.8358
0.9159	0.4661	165	0.8809	10.3857
0.8292	0.6215	220	0.7654	9.8893
0.641	0.7768	275	0.6364	9.2557
0.5445	0.9322	330	0.4931	8.6417
0.4072	1.0876	385	0.3397	8.2759
0.2378	1.2429	440	0.2414	8.1322
0.2109	1.3983	495	0.2116	7.6684
0.1641	1.5537	550	0.1940	7.6423
0.1498	1.7090	605	0.1819	7.1198
0.1445	1.8644	660	0.1752	6.8095
0.1349	2.0198	715	0.1679	6.7181
0.1032	2.1751	770	0.1661	6.7344
0.0898	2.3305	825	0.1632	6.8291
0.1032	2.4859	880	0.1606	6.7278
0.0845	2.6412	935	0.1592	6.7083
0.0958	2.7966	990	0.1578	6.5743
0.097	2.9520	1045	0.1570	6.5515
0.0689	3.1073	1100	0.1566	6.5482

Framework versions

Transformers 4.43.2
Pytorch 2.1.2
Datasets 2.20.0
Tokenizers 0.19.1