AI RESEARCH ENGINEER
Hi, I’m Noé Tits.
I build intelligent systems at the intersection of machine learning research and real-world applications.
My work spans Large Language Models, agentic AI, speech technologies, and multimodal learning. I combine deep learning expertise with practical experience solving complex problems in language, audio, and clinical data.
Applied AI at CluePoints
At CluePoints, I work on machine learning systems that detect subtle inconsistencies in clinical data. My work includes an agentic workflow that uses clinical trial protocols to guide the identification of potential data issues.
Speech technology at Flowchase
At Flowchase, I led R&D efforts in speech technology, including the implementation and fine-tuning of a wav2vec 2.0-based model for duration-aware phoneme sequence prediction. This work supported phonetic alignment, multilingual text phonetization, and speech analysis.
Research in expressive speech synthesis
During my PhD, I investigated emotionally expressive text-to-speech synthesis using a neural architecture combining a text encoder, an emotional expressiveness encoder, and an attention-based decoder. This research strengthened my expertise in neural architectures and sequence modeling.
Academic background
My academic journey included work at the University of the Basque Country’s Aholab laboratory and at UMONS’s ISIA Lab. My Master’s thesis focused on tools for pathological voice analysis, building a foundation in signal processing and voice technologies.
Across my work, I enjoy connecting foundational research with practical engineering to build intelligent systems that address real-world challenges.
EUROPEAN COMMISSION · SPEECH TECHNOLOGY
Introduction to Speech Technologies for the European Commission
I was invited by the Interinstitutional Task Force on Speech Recognition of the European Commission to introduce a workshop entitled “Speech synthesis: driver of multilingualism.”
The task force brings together representatives from EU institutions and bodies to exchange ideas and identify opportunities for collaboration in speech recognition. Its scope has expanded to AI-powered speech technologies, including speech synthesis and speech-to-speech translation.
The audience included members representing the language services of EU institutions, colleagues from the artificial intelligence network and emerging technologies committee, and representatives from academia, industry, and other stakeholders.
Workshop presentation: At the Crossroad of Expressive Speech Synthesis,
Speech Analysis, and Multi-Linguality
AI R&D @ Flowchase
PhD research subject
This presentation comes from the “Ma thèse en 180 secondes” competition aiming to explain the problematic of a PhD thesis in simple words in 2-3 minutes.