python run_eval_whisper_streaming.py \ --dataset mozilla-foundation/common_voice_11_0 \ --model_id $1 \ --split test \ --config="ar" \ --batch_size="16" \ --remove_diacritics="True" \ --language="ar" \ --device 0 \ --streaming="False"