ayertey01's picture
update model card README.md
564c5fb
metadata
license: apache-2.0
base_model: ayertey01/wav2vec2-large-xlsr-53-AsanteTwi-06second
tags:
  - generated_from_trainer
datasets:
  - common_voice_13_0
metrics:
  - wer
model-index:
  - name: wav2vec2-large-xlsr-53-AsanteTwi-07
    results:
      - task:
          name: Automatic Speech Recognition
          type: automatic-speech-recognition
        dataset:
          name: common_voice_13_0
          type: common_voice_13_0
          config: tw
          split: test
          args: tw
        metrics:
          - name: Wer
            type: wer
            value: 0.4375

wav2vec2-large-xlsr-53-AsanteTwi-07

This model is a fine-tuned version of ayertey01/wav2vec2-large-xlsr-53-AsanteTwi-06second on the common_voice_13_0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8573
  • Wer: 0.4375

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: 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: 100

Training results

Training Loss Epoch Step Validation Loss Wer
0.0656 16.67 100 0.6553 0.5
0.0354 33.33 200 0.7952 0.5
0.0329 50.0 300 0.6705 0.4375
0.0261 66.67 400 0.7253 0.4375
0.0214 83.33 500 0.8056 0.4375
0.0236 100.0 600 0.8573 0.4375

Framework versions

  • Transformers 4.31.0
  • Pytorch 2.0.1+cu118
  • Datasets 2.14.0
  • Tokenizers 0.13.3