whisper-medium-swc-drc-kat-1

This model is a fine-tuned version of openai/whisper-tiny on the Ussen/swc-drc-kat dataset. It achieves the following results on the evaluation set:

• Loss: 0.9701
• Wer Ortho: 50.0388
• Wer: 0.4938

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: constant_with_warmup
• lr_scheduler_warmup_steps: 500
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.6769	2.96	1000	0.8341	51.2296	0.5072
0.365	5.93	2000	0.8083	49.3917	0.4876
0.165	8.89	3000	0.8806	51.3073	0.5067
0.059	11.85	4000	0.9701	50.0388	0.4938

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.14.3
• Tokenizers 0.13.3