whisper-large-v2-atcosim

This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0552
• Wer: 9.9694

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
• distributed_type: multi-GPU
• num_devices: 4
• total_train_batch_size: 64
• total_eval_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 250
• training_steps: 12500

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0038	8.33	1000	0.0357	2.7829
0.001	16.67	2000	0.0384	2.0004
0.0015	25.0	3000	0.0373	31.7142
0.0001	33.33	4000	0.0437	2.3152
0.0019	41.67	5000	0.0446	7.2375
0.0	50.0	6000	0.0462	2.9033
0.0	58.33	7000	0.0490	4.3295
0.0	66.67	8000	0.0509	5.8668
0.0	75.0	9000	0.0524	7.5014
0.0	83.33	10000	0.0536	8.6405
0.0	91.67	11000	0.0546	9.5018
0.0	100.0	12000	0.0552	9.9694

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 2.0.1+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3