Edit model card
YAML Metadata Error: "model-index[0].name" is not allowed to be empty

This model was trained from scratch on the librispeech_asr dataset. It achieves the following results on the evaluation set:

  • Loss: 20.5959
  • Wer: 1.0008

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: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
6.4796 0.28 500 10.7690 1.0
6.2294 0.56 1000 10.5096 1.0
5.7859 0.84 1500 13.7547 1.0017
6.0219 1.12 2000 15.4966 1.0007
5.9142 1.4 2500 18.5919 1.0
5.6761 1.68 3000 16.9601 1.0
5.73 1.96 3500 18.9857 1.0004
4.9793 2.24 4000 18.3202 1.0007
5.2332 2.52 4500 19.5416 1.0008
4.9792 2.8 5000 20.5959 1.0008

Framework versions

  • Transformers 4.17.0.dev0
  • Pytorch 1.10.2+cu113
  • Datasets 1.18.3
  • Tokenizers 0.11.0
Downloads last month
3
Hosted inference API
or or
This model can be loaded on the Inference API on-demand.

Dataset used to train speech-seq2seq/wav2vec2-2-roberta-large-no-adapter-frozen-enc

Evaluation results

Model card error

This model's model-index metadata is invalid: Schema validation error. "[0].name" is not allowed to be empty