Whisper-Small-TF-TIMIT

This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.7104
• Wer: 98.0856

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
• 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
0.3408	3.45	500	0.3994	83.6838
0.2057	6.9	1000	0.4079	92.3470
0.0616	10.34	1500	0.5076	94.2053
0.023	13.79	2000	0.5998	95.3184
0.0043	17.24	2500	0.6825	97.1284
0.0023	20.69	3000	0.7104	98.0856

Framework versions

• Transformers 4.27.0.dev0
• Pytorch 1.13.0
• Datasets 2.1.0
• Tokenizers 0.13.2