Redve/BengaliModel

Whisper Small bn - Group 4

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2751
• Wer: 42.2126

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: 200
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.163	1.06	500	0.2064	55.6065
0.0862	2.12	1000	0.1675	47.1869
0.0475	3.18	1500	0.1696	44.8561
0.0239	4.24	2000	0.1848	43.2436
0.0119	5.3	2500	0.2081	43.5608
0.0058	6.36	3000	0.2262	43.0718
0.0024	7.42	3500	0.2427	42.4726
0.0009	8.47	4000	0.2611	42.6356
0.0005	9.53	4500	0.2709	42.3492
0.0004	10.59	5000	0.2751	42.2126

Framework versions

• Transformers 4.36.0.dev0
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0