Jarod Duret
Speech AI Scientist at AMIAD - Agence Ministérielle pour l'IA de Défense
Avignon, Provence-Alpes-Côte d'Azur, France
Summary
👤
Senior🎓
Top SchoolJarod Duret is a Speech AI Scientist with a decade of experience building and optimizing speech systems, from synthesis and translation to speaker ID and emotion recognition. Trained through a PhD and roles at Université d'Avignon and visiting stints at Johns Hopkins and NXP, he blends deep research with hands-on engineering for resource-constrained and multilingual applications. His open-source contributions to the widely used SpeechBrain toolkit include performance tuning (cuDNN benchmark mode), speech-to-speech translation integration, and K-means refactors—showing attention to both efficiency and reproducibility. Currently at France’s defence AI agency AMIAD, he applies academic rigor to real-world, large-scale speech problems. Colleagues appreciate that he pairs model-level innovation with practical deployment skills like quantization and pruning, enabling production-ready TTS and translation on embedded devices.
10 years of coding experience
5 years of employment as a software developer
Doctor of Philosophy - PhD, Artificial Intelligence, Computer Science, Doctor of Philosophy - PhD, Artificial Intelligence, Computer Science at Avignon Université
Github Skills (12)
pytorch10
machine-learning10
speech-recognition10
deep-learning10
speech-processing10
asr10
speaker-recognition9
audio-processing9
transformers9
speech-to-text8
python8
huggingface7
Programming languages (5)
Github contributions (5)
speechbrain/speechbrain
Sep 2020 - Sep 2020
A PyTorch-based Speech Toolkit
Role in this project:
ML Engineer
ML EngineerContributions:29 reviews, 6 commits, 10 PRs in 2 days
Contributions summary:Jarod primarily contributed to the SpeechBrain toolkit, a PyTorch-based framework. Their work included enabling benchmark mode in cuDNN for performance optimization, indicating a focus on efficient model execution. Further contributions involved the addition of speech-to-speech translation functionality, including integrating a CVSS recipe and related testing, demonstrating expertise in speech processing and model integration. Additionally, the user refactored parts of K-means training, and improved several test and documentation aspects.
voice-recognitionasrspeech-recognitionspeech-separationspeaker-verification
Chaanks/speechbrain
Sep 2021 - Feb 2025
A PyTorch-based Speech Toolkit
Contributions:7 PRs, 126 pushes, 15 branches in 3 years 5 months
pytorchspeech-toolkitspeech-recognitionspeech-synthesisspeech
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.