Viraj008/whisper-small-mr_v5

Whisper Small MR v5 - Viraj Patil

This model is a fine-tuned version of Viraj008/whisper-small-mr_v4 on the Common Voice 17.0, google/fleurs dataset. It achieves the following results on the evaluation set:

• Loss: 0.2496
• Wer: 34.1242

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• 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
0.0573	0.5355	1000	0.2405	36.3370
0.0314	1.0710	2000	0.2484	35.4172
0.0298	1.6064	3000	0.2410	35.1640
0.0182	2.1419	4000	0.2496	34.1242

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0