allenpoe/whisper-small-finetuned-atcosim-model-final

Whisper Base ATCOSIM

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

• Loss: 0.0493
• Wer: 8.4001

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: 1000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.5879	0.2092	100	1.3381	79.2825
0.6066	0.4184	200	0.5917	21.4916
0.1067	0.6276	300	0.1257	15.2319
0.0632	0.8368	400	0.0855	15.0973
0.0366	1.0460	500	0.0768	11.7655
0.0184	1.2552	600	0.0685	15.9992
0.0345	1.4644	700	0.0629	9.5578
0.0279	1.6736	800	0.0543	9.8607
0.0186	1.8828	900	0.0499	9.6655
0.0067	2.0921	1000	0.0493	8.4001

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1