Hugging Face's logo

ychafiqui
/
whisper-small-darija

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Moroccan Arabic
whisper
Generated from Trainer
Model card Files Metrics Community
whisper-small-darija / README.md
ychafiqui's picture
ychafiqui
End of training
427dad2 verified
preview code
|
raw
history blame contribute delete
1.49 kB
metadata
library_name: transformers
language:
  - ary
license: apache-2.0
base_model: openai/whisper-small
tags:
  - generated_from_trainer
metrics:
  - wer
model-index:
  - name: Whisper small Darija
    results: []

Whisper small Darija

This model is a fine-tuned version of openai/whisper-small on the Darija Speech to Text dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5318
  • Wer: 48.2740

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: 25
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 3
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Wer
0.4762 1.6584 1000 0.5318 48.2740

Framework versions

  • Transformers 4.45.0
  • Pytorch 2.4.1+cu121
  • Datasets 3.0.1
  • Tokenizers 0.20.0