LightFury9/whisper-tiny-DRL

whisper_tiny-drl

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

• Loss: 0.0002
• Wer: 0.0

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: 100
• training_steps: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
2.572	25.0	100	1.8923	0.0
0.2239	50.0	200	0.1132	0.0
0.0035	75.0	300	0.0022	0.0
0.0009	100.0	400	0.0008	0.0
0.0006	125.0	500	0.0005	0.0
0.0004	150.0	600	0.0004	0.0
0.0003	175.0	700	0.0003	0.0
0.0003	200.0	800	0.0003	0.0
0.0003	225.0	900	0.0002	0.0
0.0003	250.0	1000	0.0002	0.0

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0