Whisper-VAD-Small-Deep-Sparse-squeezeformer

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

• Loss: 0.4928
• Cer: 29.4300

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: 20
• 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: 2500
• training_steps: 20000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
2.999	0.7709	2500	2.9936	101.4092
2.7197	1.5418	5000	2.7733	97.9264
0.9736	2.3127	7500	1.0609	62.7590
0.5813	3.0836	10000	0.7054	42.3849
0.4856	3.8545	12500	0.5940	35.6714
0.3203	4.6253	15000	0.5318	32.1755
0.2468	5.3962	17500	0.5106	30.4677
0.1902	6.1671	20000	0.4928	29.4300

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0