whisper_sft_de

This model is a fine-tuned version of openai/whisper-small on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1840
• Wer: 28.6081

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: 8
• 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: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.2232	0.08	1000	0.2541	70.0448
0.2266	0.16	2000	0.2393	40.9830
0.22	0.24	3000	0.2278	42.1428
0.2105	0.32	4000	0.2184	41.9078
0.2108	0.4	5000	0.2090	38.9711
0.1869	0.48	6000	0.2034	28.8377
0.161	0.56	7000	0.1974	25.5598
0.1667	0.64	8000	0.1911	27.8122
0.1891	0.72	9000	0.1860	28.4944
0.1917	0.8	10000	0.1840	28.6081

Framework versions

• Transformers 4.28.0.dev0
• Pytorch 2.0.0
• Datasets 2.11.0
• Tokenizers 0.13.3