xbilek25/whisper-small-train-v2.1

basic_train_basic_test 1000 similar params: per_device_train_batch_size=32, # bylo 16 a pod tim 1 gradient_accumulation_steps=2, warmup_steps=300, max_steps=3000

This model is a fine-tuned version of openai/whisper-small on the xbilek25/xbilek25/train_set_1000_de_en_de dataset. It achieves the following results on the evaluation set:

• Loss: 0.3813
• Wer: 16.0452

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: 500
• training_steps: 3000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0113	7.02	500	0.3458	17.0373
0.0015	15.02	1000	0.3461	18.2005
0.0007	23.02	1500	0.3652	16.2504
0.0005	31.02	2000	0.3741	16.3531
0.0004	39.01	2500	0.3790	15.6688
0.0004	47.01	3000	0.3813	16.0452

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2