Mahmoud Ashraf
Machine Learning Engineer at Tarteel AI
Faiyum, Egypt
Summary
🤩
Rockstar🎓
Top SchoolMahmoud Ashraf is a Machine Learning Engineer with eight years of experience specializing in speech recognition and deep learning, currently building ML solutions at Tarteel AI. He has a strong open-source footprint, contributing to notable ASR projects like whisper-diarization and whisperX by adding multilingual support, word-level alignment, and diarization integrations. His background spans research and applied roles—FAO, Unify, and top-rated Upwork engagements—where he delivered production-ready models and cross-framework tooling. Mahmoud’s engineering breadth includes backend ML framework work (TensorFlow/PyTorch/Paddle) and practical improvements such as Windows path handling and Hugging Face integrations, reflecting a knack for bridging research models to real-world systems. Trained in communications and electrical engineering in Egypt, he combines academic rigor with hands-on systemization of speech ML pipelines.
8 years of coding experience
3 years of employment as a software developer
Bachelor of Engineering - BE, Communications and Information Engineering, Very Good, Bachelor of Engineering - BE, Communications and Information Engineering, Very Good at Zewail City of Science and TechnologyBachelor of Engineering - BE, Electrical and Electronics Engineering, Very Good, Bachelor of Engineering - BE, Electrical and Electronics Engineering, Very Good at Fayoum University
Arabic, English
Github Skills (18)
speaker-diarization10
ivy10
speech-to-text10
pytorch10
python10
machine-learning10
huggingface10
tensorflow10
natural-language-processing10
asr10
speech-recognition10
nlp10
deeplearning-ai9
deep-learning9
git8
Programming languages (6)
Github contributions (5)
MahmoudAshraf97/whisper-diarization
Jan 2023 - Mar 2023
Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper
Role in this project:
ML Engineer
ML EngineerContributions:19 reviews, 28 commits, 27 PRs in 1 month
Contributions summary:Mahmoud primarily contributed to the development of an automatic speech recognition (ASR) system with speaker diarization based on OpenAI Whisper, as indicated by the repository description and topics. Their work involved integrating the Whisper model for transcription, aligning word-level timestamps, and integrating NeMo for speaker diarization. They also added multilingual support, incorporated punctuation restoration, and implemented path handling for Windows.
asrspeaker-diarizationspeechspeech-recognitionspeech-to-text
m-bain/whisperX
Jan 2023 - Feb 2023
WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
Role in this project:ML Engineer
ML EngineerContributions:8 commits, 8 PRs, 27 comments in 6 days
Contributions summary:Mahmoud contributed to the project by integrating Hugging Face models and resources. This includes adding the ability to use a Hugging Face access token for PyAnnote models. Furthermore, the user added several wav2vec2 models for speech alignment, including models for Arabic, Russian, Polish, Hungarian, Finnish, Persian, Greek, and Turkish. These additions expanded the project's capabilities to support multiple languages for automatic speech recognition.
speaker-diarizationasrwhisperspeech-to-textrecognition
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.