whisper-base-synth-v1

This model is a fine-tuned version of ihanif/whisper-base-cv20-v2 on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

• eval_loss: 1.4352
• eval_wer_ortho: 816.3428
• eval_wer: 837.2591
• eval_runtime: 1020.837
• eval_samples_per_second: 2.744
• eval_steps_per_second: 0.043
• epoch: 3.1915
• step: 300

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100
• mixed_precision_training: Native AMP

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.21.1