dgx1_whisper_base_mozilla_noisy_teacher_distil_epochs_50_batch_16

This model is a fine-tuned version of rohitp1/dgx1_whisper_base_finetune_teacher_babble_noise_mozilla_100_epochs_batch_16 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0845
• Wer: 33.7337

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 256
• total_train_batch_size: 4096
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.2
• num_epochs: 50
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1214	4.41	150	0.5017	32.6283
0.1796	8.82	300	0.7222	33.1175
0.3176	13.23	450	0.8749	33.3612
0.3986	17.64	600	0.9770	33.5457
0.4497	22.06	750	1.0385	33.7111
0.4816	26.47	900	1.0845	33.7337

Framework versions

• Transformers 4.25.1
• Pytorch 1.12.1
• Datasets 2.8.0
• Tokenizers 0.13.2