AudioCourseU5-ASR

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

• Loss: 0.6438
• Wer Ortho: 34.4849
• Wer: 0.3406

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: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.3065	3.57	100	0.4921	36.8908	0.3577
0.0391	7.14	200	0.5425	35.3486	0.3436
0.0042	10.71	300	0.5878	35.6570	0.3495
0.0012	14.29	400	0.6206	34.2998	0.3377
0.0007	17.86	500	0.6438	34.4849	0.3406

Framework versions

• Transformers 4.32.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3