wav2vec2-ft-lre6

This model is a fine-tuned version of jonatasgrosman/wav2vec2-large-english on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8832
• Wer: 1.0
• Cer: 1.0

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-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 10000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer	Cer
1.006	0.4165	1000	0.9763	1.0	1.0
0.9162	0.8330	2000	0.9393	1.0	1.0
0.7996	1.2495	3000	0.9507	1.0	1.0
0.8934	1.6660	4000	0.8868	1.0	1.0
0.8691	2.0825	5000	0.8926	1.0	1.0
0.8369	2.4990	6000	0.8859	1.0	1.0
0.8373	2.9155	7000	0.8960	1.0	1.0
0.8742	3.3319	8000	0.8814	1.0	1.0
0.801	3.7484	9000	0.8832	1.0	1.0
0.7813	4.1649	10000	0.8832	1.0	1.0

Framework versions

• Transformers 4.52.3
• Pytorch 2.5.1
• Datasets 3.6.0
• Tokenizers 0.21.1