IoanaLiviaPopescu/synth-data-v1-whisper-small

This model is a fine-tuned version of openai/whisper-small on the IoanaLiviaPopescu/SyntheticVoiceHoroscope-v1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4825
• Wer: 19.8230

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_BNB 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: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
No log	0	0	0.6024	27.8812
0.2529	1.0	13	0.5140	20.6343
0.1068	2.0	26	0.4849	20.0996
0.0609	3.0	39	0.4808	19.9521
0.0427	4.0	52	0.4825	19.8230
0.0349	5.0	65	0.4850	19.8783

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1