Clément Pages
Ingénieur D'études En Traitement Automatique De La Parole
Toulouse, Occitania, France
Summary
🤩
Rockstar🎓
Top SchoolClément Pages is a French AI engineer specializing in speech processing and robotic systems, with four years of experience building deep learning models for speaker segmentation, clustering and source separation. Based in Toulouse, he contributes to the well-regarded open-source pyannote.audio project, where he implemented multi-speaker statistics pooling and caching features to improve diarization workflows. Employed at IRIT on a DGA-linked SeRiouSLy project, he develops multitask models that jointly segment, cluster and separate speakers for real-world audio scenarios. Trained as an engineer in Robotic and Interactive Systems, he also has hands-on embedded-systems experience integrating mmWave radar for obstacle detection on mobile robots. Pragmatic and research-minded, he combines academic rigor with production-oriented contributions to open-source ML tooling.
4 years of coding experience
Diplôme d'ingénieur, Systèmes Robotiques et Interactifs, Diplômé en Systèmes Robotiques et Interactifs, Diplôme d'ingénieur, Systèmes Robotiques et Interactifs, Diplômé en Systèmes Robotiques et Interactifs at Université Paul Sabatier Toulouse III - UPSSITECH école d'ingénieurs interneCycle Universitaire de Préparation aux Grandes Ecoles, Sciences fondamentales (mathématiques, physique, informatique), Cycle Universitaire de Préparation aux Grandes Ecoles, Sciences fondamentales (mathématiques, physique, informatique) at Université Paul Sabatier Toulouse IIIBaccalauréat scientifique, Spécialité Informatique et Sciences du Numérique, Baccalauréat obtenu mention Très Bien, Baccalauréat scientifique, Spécialité Informatique et Sciences du Numérique, Baccalauréat obtenu mention Très Bien at Lycée Déodat de Séverac Toulouse
English, German, Spanish, French
Github Skills (12)
neural-network10
speaker-diarization10
model-building10
pytorch10
speech-processing10
modeling10
model-driven10
model-driven-development10
pytorch-lightning9
python7
testing7
pre-trained-model6
Programming languages (5)
Github contributions (5)
pyannote/pyannote-audio
Sep 2022 - Apr 2025
Neural building blocks for speaker diarization: speech activity detection, speaker change detection, overlapped speech detection, speaker embedding
Role in this project:
ML Engineer
ML EngineerContributions:25 reviews, 26 PRs, 55 comments in 2 years 6 months
Contributions summary:Clément primarily contributed to the implementation and testing of machine learning model components within the pyannote-audio framework. Their work includes adding support for multi-speaker statistics pooling, enhancing the model's ability to handle speaker diarization tasks. They also added options for caching training metadata. The user's contributions directly relate to the core functionalities of the repository, involving code modifications and test additions.
diarizationspeech-recognitionspeaker-verificationspeech-processingspeaker-diarization
SkyEyeWeather/SunBot
May 2023 - Nov 2024
Sunbot is a discord bot that provides commands to get current and forecast weather using the Visual Crossing API
Contributions:2 releases, 10 reviews, 63 PRs in 1 year 6 months
botdiscord-botpython3visualcrossingweather
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.