anuragshas/whisper-large-v2-lv

Whisper Large-v2 Latvian

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

• Loss: 0.2634
• Wer: 17.7799

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-06
• 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: 100
• training_steps: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2124	3.02	200	0.2485	18.4061
0.0704	6.05	400	0.2634	17.7799
0.0379	10.01	600	0.3103	17.8178
0.0228	13.03	800	0.3555	18.4061
0.0139	16.06	1000	0.3733	18.4535

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu117
• Datasets 2.7.1.dev0
• Tokenizers 0.13.2