halima12378/whisper-small-colab

Whisper Small Hi - Sanchit Gandhi

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: 0.3001
• Wer: 8.9027

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 50
• training_steps: 500
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2016	0.13	100	0.3489	10.7292
0.2573	0.27	200	0.3331	10.8477
0.268	0.4	300	0.3292	10.0251
0.2338	0.53	400	0.3094	9.3558
0.1767	0.67	500	0.3001	8.9027

Framework versions

• Transformers 4.39.3
• Pytorch 2.7.1+cu126
• Datasets 3.6.0
• Tokenizers 0.15.2