---

language:

• ar
• arz license: apache-2.0 base_model: openai/whisper-large-v3-turbo library_name: transformers pipeline_tag: automatic-speech-recognition tags:
• automatic-speech-recognition
• speech
• egyptian-arabic
• whisper
• whisper.cpp
• ggml model-index:
• name: Drone Transcripter 4 results:
  • task: { type: automatic-speech-recognition, name: Speech Recognition } dataset: { type: egyptian-lectures, name: Egyptian school lectures (held-out, 600 clips) } metrics: [{ type: wer, value: 10.2, name: WER }]
  • task: { type: automatic-speech-recognition, name: Speech Recognition } dataset: { type: mgb3, name: MGB-3 Egyptian (dev subset, 773 clips, multi-reference) } metrics: [{ type: wer, value: 15.3, name: MR-WER }]
  • task: { type: automatic-speech-recognition, name: Speech Recognition } dataset: { type: fleurs, name: FLEURS ar_eg (MSA) } metrics: [{ type: wer, value: 10.5, name: WER }]
  • task: { type: automatic-speech-recognition, name: Speech Recognition } dataset: { type: gigaspeechbench, name: GigaSpeechBench EGY } metrics: [{ type: wer, value: 48.46, name: WER (official scorer) }]

---

Drone Transcripter 4

809M-parameter Whisper large-v3-turbo, fully fine-tuned by Drone Academy for Egyptian Arabic: classroom lectures first, then everyday speech, Modern Standard Arabic and Arabic/English code-switching. Trained on 329.5 h of human-reviewed and Gemini-cleaned Egyptian audio (75,390 clips) in two epochs on one A100.

Try it live: https://asr.drone-academy.online.

Results (WER, lower is better)

set	this model	base large-v3-turbo	notes
Egyptian lectures, held-out (600 clips)	10.2	64.0	our target domain
MGB-3 Egyptian, multi-reference (773 clips)	15.3		GigaSpeech 3 scores 17.99 on the same clips
FLEURS ar_eg (MSA)	10.5
Casablanca (Egyptian)	40.2		everyday speech; GigaSpeech 3 38.1
GigaSpeechBench EGY	48.46		official scorer, with punctuation
English words inside Arabic speech	90.1 % correct
Numbers	91.7 % correct

Files

file	use
drone-transcripter-v4/	Hugging Face transformers checkpoint (safetensors), WhisperForConditionalGeneration
ggml-drone-transcripter-v4-f16.bin	whisper.cpp model (f16), runs on CPU or Metal/CUDA via whisper.cpp
ggml-drone-transcripter-v4-q5_0.bin	whisper.cpp model quantised to q5_0 (574 MB); same accuracy class, 2.8× smaller, the one to download for laptops

Use with transformers

from transformers import pipeline
asr = pipeline("automatic-speech-recognition", model="AbdelrhmanX2/drone-transcripter-4", device="cuda")
print(asr("clip.wav", generate_kwargs={"language": "ar", "task": "transcribe"})["text"])

Use with whisper.cpp

whisper-cli -m ggml-drone-transcripter-v4-q5_0.bin -l ar -f clip.wav

Training

Base openai/whisper-large-v3-turbo, full fine-tune, 2 epochs, 4,712 steps, batch 32, lr 1e-5 with 5 % warm-up and linear decay, bf16, one A100-40GB (~3.5 h). Data: 122.4 h of human-reviewed lectures plus 207.1 h of platform lectures re-cut and cleaned with Gemini. Songs and music excluded.

Limitations

Strongest on lectures and classroom speech. Casual conversation and noisy phone audio are weaker (Casablanca row). Outputs are unpunctuated Egyptian Arabic in Arabic script; English words are kept in Latin script.

Licence and citation

Apache-2.0, matching the base model. Please cite Drone Academy and link to this page.