zuazo/whisper-base-ca

Whisper Base Catalan

This model is a fine-tuned version of openai/whisper-base on the mozilla-foundation/common_voice_13_0 ca dataset. It achieves the following results on the evaluation set:

• Loss: 0.2782
• Wer: 13.7897

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: 2.5e-05
• train_batch_size: 128
• eval_batch_size: 64
• 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
0.0975	3.05	1000	0.3560	19.4421
0.1381	7.04	2000	0.3066	16.1486
0.1302	11.04	3000	0.2902	15.4296
0.1089	15.03	4000	0.2699	14.0726
0.0505	19.03	5000	0.2782	13.7897

Framework versions

• Transformers 4.33.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.14.4
• Tokenizers 0.13.3