wh_medium_fl_no_langid_fleurs_47716_trial

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

• Loss: 0.2134
• Global Wer: 24.2213

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-06
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 50
• training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Global Wer
0.7885	0.0945	50	0.6567	39.1274
0.4267	0.1890	100	0.3318	33.3104
0.2592	0.2836	150	0.2767	30.4412
0.2661	0.3781	200	0.2612	28.7806
0.2409	0.4726	250	0.2465	29.5568
0.2190	0.5671	300	0.2373	27.2477
0.2223	0.6616	350	0.2251	26.1079
0.2042	0.7561	400	0.2216	24.8305
0.1849	0.8507	450	0.2146	24.0739
0.1964	0.9452	500	0.2134	24.2213

Framework versions

• Transformers 5.0.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 3.6.0
• Tokenizers 0.22.2