Malak2003m
/

whisper-small-arz-en-codeswitching

Automatic Speech Recognition

egyptian-arabic

knowledge-distillation

Generated from Trainer

Model card Files Files and versions

Metrics Training metrics Community

Whisper Small Egyptian Arabic–English Code-Switching

This model is a fine-tuned version of openai/whisper-small on the FLEURS Egyptian Arabic (ar_eg) dataset.

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: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
training_steps: 4000
mixed_precision_training: Native AMP

Framework versions

Transformers 5.8.0.dev0
Pytorch 2.10.0+cu128
Datasets 4.8.5
Tokenizers 0.22.2

Downloads last month: 23

Safetensors

Model size

0.2B params

Tensor type

F32

·

Model tree for Malak2003m/whisper-small-arz-en-codeswitching

Base model

openai/whisper-small

Finetuned

(3557)

this model

Dataset used to train Malak2003m/whisper-small-arz-en-codeswitching