rse-mfm/whisper-large-v3-mfm

Whisper Large V3 - MFM

This model is a fine-tuned version of openai/whisper-large-v3 on the PX Corpus dataset. It achieves the following results on the evaluation set:

• Loss: 0.3224
• Wer: 8.5976

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: 8
• 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_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0035	14.0845	1000	0.2718	11.4635
0.0001	28.1690	2000	0.2975	8.5010
0.0002	42.2535	3000	0.3095	8.6137
0.0	56.3380	4000	0.3224	8.5976

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0