nikhilbh/whisper-medium-custom-hi

Whisper Medium Custom Hi - Nikhil Bhargava

This model is a fine-tuned version of openai/whisper-medium on the mozilla-foundation/common_voice_11_0 hi dataset. It achieves the following results on the evaluation set:

• Loss: 0.3972
• Wer: 0.2487

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: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• 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
0.0282	4.89	1000	0.2700	0.2647
0.0025	9.78	2000	0.3434	0.2554
0.0001	14.67	3000	0.3640	0.2471
0.0	19.56	4000	0.3902	0.2494
0.0	24.45	5000	0.3972	0.2487

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3