verbalex-hi

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

• Loss: 0.0554
• Wer: 1.6019

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: 16
• eval_batch_size: 8
• seed: 42
• 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
0.0011	5.0505	1000	0.0504	1.7762
0.0001	10.1010	2000	0.0529	1.6268
0.0001	15.1515	3000	0.0541	1.6434
0.0001	20.2020	4000	0.0550	1.6102
0.0	25.2525	5000	0.0554	1.6019

Framework versions

• Transformers 4.40.2
• Pytorch 2.1.2
• Datasets 2.16.0
• Tokenizers 0.19.1