CheeLi03/whisper-base-tr-8

Whisper Base Turkish 8000 - Chee Li

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

• Loss: 0.5649
• Wer: 25.8479

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 850
• training_steps: 8000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.1634	5.5866	1000	0.4092	24.8833
0.0075	11.1732	2000	0.4509	24.2066
0.0024	16.7598	3000	0.4874	24.1910
0.0012	22.3464	4000	0.5125	24.3777
0.0008	27.9330	5000	0.5305	24.5644
0.0005	33.5196	6000	0.5473	24.8289
0.0004	39.1061	7000	0.5592	24.9922
0.0003	44.6927	8000	0.5649	25.8479

Framework versions

• Transformers 4.43.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1