Hugging Face's logo

makhataei
/
Whisper-Small-Common-Voice

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Persian
whisper
fa-asr
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Whisper-Small-Common-Voice / README.md
makhataei's picture
makhataei
End of training
8df4d03 verified
preview code
|
raw
history blame
2.15 kB
metadata
language:
  - fa
license: apache-2.0
base_model: makhataei/Whisper-Small-Common-Voice
tags:
  - fa-asr
  - generated_from_trainer
datasets:
  - mozilla-foundation/common_voice_17_0
metrics:
  - wer
model-index:
  - name: Whisper Small Persian
    results: []

Whisper Small Persian

This model is a fine-tuned version of makhataei/Whisper-Small-Common-Voice on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7219
  • Wer: 46.5235

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-06
  • train_batch_size: 10
  • eval_batch_size: 10
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 40
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 50
  • training_steps: 1000

Training results

Training Loss Epoch Step Validation Loss Wer
0.0057 0.15 100 0.7193 45.7885
0.0051 0.31 200 0.7189 44.8246
0.0027 0.46 300 0.7138 46.2762
0.0025 0.61 400 0.7172 45.8786
0.0029 0.77 500 0.7175 47.2909
0.0071 0.92 600 0.7199 46.7847
0.0273 1.07 700 0.7144 46.7616
0.0015 1.23 800 0.7174 46.6391
0.0016 1.38 900 0.7211 46.4958
0.0013 1.53 1000 0.7219 46.5235

Framework versions

  • Transformers 4.35.2
  • Pytorch 2.0.1+cu117
  • Datasets 2.15.0
  • Tokenizers 0.15.0