wav2vec2-base-960h-timit-demo-colab

This model is a fine-tuned version of facebook/wav2vec2-base-960h on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2002
• Wer: 0.2160

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: 0.0001
• train_batch_size: 32
• 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: 1000
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
5.7805	4.0	500	3.0558	1.0
2.2936	8.0	1000	0.2937	0.3479
0.4155	12.0	1500	0.2108	0.2473
0.2439	16.0	2000	0.2313	0.2391
0.1617	20.0	2500	0.2003	0.2255
0.1443	24.0	3000	0.2175	0.2207
0.119	28.0	3500	0.2002	0.2160

Framework versions

• Transformers 4.11.3
• Pytorch 1.10.0+cu111
• Datasets 1.18.3
• Tokenizers 0.10.3