peterwei89/whisper-small-hi-new

Whisper Small hindi - Peter Wei

This model is a fine-tuned version of openai/whisper-small on the hindi_project1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4301
• Wer: 34.1361

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
0.082	2.44	1000	0.2958	35.3213
0.0221	4.89	2000	0.3454	33.7806
0.0013	7.33	3000	0.4056	34.1531
0.0005	9.78	4000	0.4301	34.1361

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0