wh_small_cv_no_langid_common_voices_47716_trial

This model is a fine-tuned version of openai/whisper-small on the common_voice_22_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3836
• Global Wer: 49.6390

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
0.7614	0.0411	50	0.6498	65.3871
0.4647	0.0821	100	0.5258	58.4388
0.4257	0.1232	150	0.4795	55.3935
0.3575	0.1642	200	0.4472	52.4448
0.3408	0.2053	250	0.4421	51.8658
0.2971	0.2463	300	0.4174	50.6684
0.3103	0.2874	350	0.4073	49.4067
0.2942	0.3284	400	0.3930	49.5139
0.2978	0.3695	450	0.3835	48.4202
0.2892	0.4105	500	0.3836	49.6390

Framework versions

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