BlueRaccoon/whisper-medium-br

Whisper Medium Breton

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

• Loss: 0.8486
• Wer: 41.6117

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: 4e-06
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 400
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0602	5.03	1000	0.7324	43.6957
0.0036	10.05	2000	0.8486	41.6117
0.001	15.08	3000	0.9033	42.0458
0.0004	20.1	4000	0.9351	41.6811
0.0003	25.13	5000	0.9468	41.7853

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.1+cu117
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2