EdoAbati/whisper-large-v2-it

Whisper Large v2 Italian

This model is a fine-tuned version of openai/whisper-large-v2 on the mozilla-foundation/common_voice_11_0 it dataset. It achieves the following results on the evaluation set:

• Loss: 0.1332
• Wer: 4.5576

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: 16
• seed: 42
• distributed_type: multi-GPU
• gradient_accumulation_steps: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 6000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1684	0.17	1000	0.1620	6.4620
0.1174	0.33	2000	0.1418	5.5663
0.069	1.1	3000	0.1400	5.2865
0.0649	1.27	4000	0.1315	4.8932
0.0334	2.04	5000	0.1368	4.6845
0.037	2.21	6000	0.1332	4.5576

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.1+cu117
• Datasets 2.8.1.dev0
• Tokenizers 0.13.2