whisper-medium-es / README.md
juancopi81's picture
Librarian Bot: Add base_model information to model (#1)
5554c51
metadata
language:
  - es
license: apache-2.0
tags:
  - whisper-event
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_11_0
metrics:
  - wer
base_model: openai/whisper-medium
model-index:
  - name: Whisper Medium Es - Juan Carlos Piñeros
    results:
      - task:
          type: automatic-speech-recognition
          name: Automatic Speech Recognition
        dataset:
          name: Common Voice 11.0
          type: mozilla-foundation/common_voice_11_0
          config: es
          split: test
          args: es
        metrics:
          - type: wer
            value: 5.421819787985865
            name: Wer

Whisper Medium Es - Juan Carlos Piñeros

This model is a fine-tuned version of openai/whisper-medium on the Common Voice 11.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1672
  • Wer: 5.4218

Using the script provided in the Whisper Sprint (Dec. 2022) the models achieves these results on the evaluation sets (WER):

  • google/fleurs: 5.88
  • mozilla-foundation/common_voice_11_0: XXX

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: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • training_steps: 3000
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.0792 0.33 1000 0.1904 6.0493
0.0851 0.67 2000 0.1757 5.9558
0.0946 1.0 3000 0.1672 5.4218

Framework versions

  • Transformers 4.26.0.dev0
  • Pytorch 1.12.1+cu113
  • Datasets 2.7.1.dev0
  • Tokenizers 0.13.2