whisper-tiny-en-us-minds14

This model is a fine-tuned version of openai/whisper-tiny on the PolyAI/minds14 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.3247
• eval_wer_ortho: 36.7057
• eval_wer: 36.6588
• eval_runtime: 20.5021
• eval_samples_per_second: 5.512
• eval_steps_per_second: 0.39
• epoch: 107.14
• step: 3000

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: 4000

Framework versions

• Transformers 4.35.1
• Pytorch 2.1.0
• Datasets 2.14.6
• Tokenizers 0.14.1