deepdml/whisper-medium-gl-cv17

Whisper Medium gl

This model is a fine-tuned version of openai/whisper-medium on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1998
• Wer: 5.3774

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: 32
• 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: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1504	0.9099	1000	0.1736	6.9625
0.0845	1.8198	2000	0.1567	5.8877
0.0351	2.7298	3000	0.1649	5.7172
0.0099	3.6397	4000	0.1853	5.4799
0.0037	4.5496	5000	0.1998	5.3774

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1