danielizham/whisper-small-es

Whisper Small Spanish

This model is a fine-tuned version of openai/whisper-small on the mozilla-foundation/common_voice_11_0 es dataset. It achieves the following results on the evaluation set:

• Loss: 0.2210
• Wer: 8.2123

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 5000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1386	4.01	1000	0.2464	9.8000
0.1098	8.01	2000	0.2272	8.6229
0.028	12.02	3000	0.2577	8.6956
0.1083	16.02	4000	0.2210	8.2123
0.0189	20.03	5000	0.2520	8.4455

Framework versions

• Transformers 4.29.0.dev0
• Pytorch 2.0.0+cu117
• Datasets 2.12.1.dev0
• Tokenizers 0.13.3