metadata

library_name: transformers
license: apache-2.0
base_model: facebook/data2vec-audio-base-960h
tags:
  - generated_from_trainer
datasets:
  - gigaspeech
metrics:
  - wer
model-index:
  - name: wav2vec_5e-5_3
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: gigaspeech
          type: gigaspeech
          config: xs
          split: validation
          args: xs
        metrics:
          - name: Wer
            type: wer
            value: 0.29402661714639433

wav2vec_5e-5_3

This model is a fine-tuned version of facebook/data2vec-audio-base-960h on the gigaspeech dataset. It achieves the following results on the evaluation set:

Loss: 0.6689
Wer: 0.2940

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: 5e-05
train_batch_size: 1
eval_batch_size: 32
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 8
optimizer: Use OptimizerNames.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: 200
num_epochs: 3
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
1.1835	1.0	1174	0.6329	0.3020
1.2218	2.0	2348	0.6741	0.2961
0.4211	2.9978	3519	0.6689	0.2940

Framework versions

Transformers 4.50.3
Pytorch 2.6.0+cu124
Datasets 3.5.0
Tokenizers 0.21.1