HuBERT-base-libriSpeech-demo-colab

This model is a fine-tuned version of facebook/hubert-large-ls960-ft on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1456
• Wer: 0.2443

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: 2
• 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
7.6395	13.51	500	3.1933	0.9930
2.5994	27.03	1000	0.1456	0.2443

Framework versions

• Transformers 4.20.1
• Pytorch 1.11.0+cu113
• Datasets 2.3.2
• Tokenizers 0.12.1