README.md

metadata

library_name: transformers
license: apache-2.0
base_model: lgris/bp400-xlsr
tags:
  - generated_from_trainer
datasets:
  - common_voice_13_0
metrics:
  - wer
model-index:
  - name: wav2vec2-large-xlsr-tutorial-pt-br-5.0
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_13_0
          type: common_voice_13_0
          config: pt
          split: None
          args: pt
        metrics:
          - name: Wer
            type: wer
            value: 0.7539496503496502

wav2vec2-large-xlsr-tutorial-pt-br-5.0

This model is a fine-tuned version of lgris/bp400-xlsr on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:

• Loss: 0.6661
• Wer: 0.7539

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: 200
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
5.0104	6.3492	200	1.2945	0.9866
0.7801	12.6984	400	0.6661	0.7539

Framework versions

• Transformers 4.46.0.dev0
• Pytorch 2.4.1+cu121
• Datasets 3.0.1.dev0
• Tokenizers 0.20.0