metadata

base_model: daila/wav2vec2-large-xls-r-300m-vi-colab
tags:
  - generated_from_trainer
datasets:
  - common_voice_16_1
metrics:
  - wer
model-index:
  - name: wav2vec2-large-xls-r-300m-vi-colab
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_16_1
          type: common_voice_16_1
          config: vi
          split: test
          args: vi
        metrics:
          - name: Wer
            type: wer
            value: 0.5894672631150875

wav2vec2-large-xls-r-300m-vi-colab

This model is a fine-tuned version of daila/wav2vec2-large-xls-r-300m-vi-colab on the common_voice_16_1 dataset. It achieves the following results on the evaluation set:

Loss: 1.6432
Wer: 0.5895

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: 16
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 30
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
0.0916	4.52	400	1.5440	0.6357
0.1344	9.04	800	1.6043	0.6543
0.0926	13.56	1200	1.7226	0.6365
0.0703	18.08	1600	1.5989	0.6048
0.0557	22.6	2000	1.6714	0.6001
0.051	27.12	2400	1.6432	0.5895

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1