whisper-tiny-it

This model is a fine-tuned version of openai/whisper-tiny on the Common Voice 11 dataset.

Language: Italian

It achieves the following results on the evaluation set:

• Loss: 0.3958
• Wer: 26.61

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: 32
• seed: 42
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 256
• total_eval_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: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3997	1.68	1000	0.4651	0.3038
0.3026	3.35	2000	0.4086	0.2743
0.2874	5.03	3000	0.3958	0.2661

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2