wav2vec2-finetuned-cv

This model is a fine-tuned version of facebook/wav2vec2-large-960h on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1134
• Wer: 0.0535

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: 0.0003
• train_batch_size: 512
• eval_batch_size: 512
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 50
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.3111	0.5464	100	0.2240	0.0993
0.281	1.0929	200	0.1988	0.0899
0.206	1.6393	300	0.1538	0.0714
0.1653	2.1858	400	0.1309	0.0623
0.1334	2.7322	500	0.1134	0.0535

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.22.1