Julian Weber
Senior Deep Learning Researcher at Cantina
Paris, Ile-de-France
Summary
🤩
Rockstar🎓
Top SchoolJulian Weber is a Senior Deep Learning Researcher based in Paris with nine years of experience building production-ready generative AI and speech systems. He has driven neural TTS and expressive speech work at Coqui and now leads deep learning research at Cantina, combining hands-on model development with backend and DevOps expertise. His open-source contributions include audio resampling tooling, French text-cleaning and training improvements in widely used TTS toolkits such as Mozilla TTS and Coqui TTS (24k-star community), showing attention to both research and production quality. Earlier roles span few-shot voice and facial animation, transformer fine-tuning for NLP, and full-stack engineering for accessibility and startup products. He pairs academic training in engineering and big data with practical experience shipping voice tech in low-data scenarios. Colleagues describe him as a lifelong learner who surfaces pragmatic solutions that bridge research prototypes and scalable deployments.
9 years of coding experience
6 years of employment as a software developer
Master of Engineering - MEng computer science, Master of Engineering - MEng computer science at ESILV - Ecole Supérieure d'Ingénieurs Léonard de VinciBigData Managment and Analytics, BigData Managment and Analytics at Griffith College DublinPrépa PSI* Lycée Franklin Roosevelt
Stackoverflow
51reputation
3kreached
1answer
2questions
Github Skills (25)
pytorch10
python10
text-to-speech10
audio-processing10
gtts10
freetts10
deeplearning-ai10
deep-learning10
tensorflow10
librosa10
testing9
machine-learning9
bash9
loss-functions9
refactor7
Programming languages (8)
Github contributions (5)
coqui-ai/TTS
Mar 2021 - Jan 2023
🐸💬 - a deep learning toolkit for Text-to-Speech, battle-tested in research and production
Role in this project:
Backend & DevOps Engineer
Backend & DevOps EngineerContributions:4 releases, 91 reviews, 172 commits in 1 year 10 months
Contributions summary:Julian primarily contributed to the development of a resampling script and associated testing infrastructure within the TTS project. This involved writing Python code using libraries like librosa for audio processing, creating a shell script for running tests, and setting up test cases to validate the resampling functionality. Additionally, the user made improvements to the project's text cleaning components, and implemented linter fixes.
text-to-speechpythonspeaker-encodingsasrglow-tts
mozilla/TTS
Sep 2020 - Sep 2020
:robot: :speech_balloon: Deep learning for Text to Speech (Discussion forum: https://discourse.mozilla.org/c/tts)
Role in this project:Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:5 commits, 2 PRs, 99 comments in 7 days
Contributions summary:Julian contributed to the Mozilla TTS repository by addressing bugs related to pickling and module renaming. They added and improved text cleaning functionality, specifically for French, including abbreviation expansion. Furthermore, the user integrated and modified loss functions, and implemented adjustments in the training process. The user's work included changes to the dataset split and the integration of a new config check function.
balloonpythondataset-analysisglow-ttsspeech-recognition
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.