mn367/whisper-medium-1hr

Whisper Medium 1hr

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

• eval_loss: 0.6053
• eval_wer: 17.5426
• eval_runtime: 651.6941
• eval_samples_per_second: 2.363
• eval_steps_per_second: 0.296
• epoch: 37.5
• step: 900

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-06
• 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: 1600
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.26.0.dev0
• Pytorch 1.13.0+cu116
• Datasets 2.8.0
• Tokenizers 0.13.2