xbilek25/w-m-lang_de-set_en
04
<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
model trenovan na en setu, nastaveni jazyka de
This model is a fine-tuned version of openai/whisper-medium on the xbilek25/trainset1st1000deende dataset. It achieves the following results on the evaluation set:
- Loss: 0.2223
- Wer: 53.5483
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
- trainbatchsize: 16
- evalbatchsize: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lrschedulertype: linear
- lrschedulerwarmup_steps: 1
- training_steps: 2000
- mixedprecisiontraining: Native AMP
Training results
Framework versions
- Transformers 4.37.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.15.2