whisper-tiny-full-data-v4

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

• Loss: 0.1781

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: 0.00015
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• distributed_type: tpu
• num_devices: 8
• total_train_batch_size: 256
• total_eval_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5000
• training_steps: 63840

Training results

Training Loss	Epoch	Step	Validation Loss
0.3932	1.57	5000	0.5166
0.2683	3.13	10000	0.3819
0.2227	4.7	15000	0.3289
0.1905	6.27	20000	0.2941
0.1691	7.83	25000	0.2677
0.1421	9.4	30000	0.2499
0.1355	10.96	35000	0.2305
0.1126	12.53	40000	0.2185
0.0954	14.1	45000	0.2061
0.0858	15.66	50000	0.1951
0.0715	17.23	55000	0.1880
0.0634	18.8	60000	0.1808

Framework versions

• Transformers 4.28.1
• Pytorch 2.1.0a0+gitcc01568
• Datasets 2.12.0
• Tokenizers 0.13.3