whisper-small-en-hi

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

• Loss: 0.3279
• Wer: 24.0479

Model description

Two datasets are used for two different languages, for hindi mozilla-foundation/common_voice_11_0 is used and for english google/fleurs is used. with combination of two dataset wer has decreased significantly.

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: 3000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.059	2.52	1500	0.2881	24.7722
0.0084	5.03	3000	0.3279	24.0479

Framework versions

• Transformers 4.33.0
• Pytorch 1.11.0+cu113
• Datasets 2.1.0
• Tokenizers 0.13.3