Toumi Bouchentouf

Papers

Total Citations

H-Index

About

Toumi Bouchentouf is a researcher at the intersection of speech recognition, robotics, and artificial intelligence, with a focus on making human-robot interaction more seamless and robust. His work addresses critical challenges in deploying AI systems in real-world environments, particularly for assistant robots. Bouchentouf’s most cited paper, “Adapting Off-the-Shelf Speech Recognition Systems for Novel Words” (2023, 6 citations), tackles the persistent problem of Out-of-Vocabulary (OOV) words—such as proper nouns and new terms—that cause dialog systems to fail. By proposing methods to adapt existing speech recognition models, he offers a practical path to more resilient voice interfaces. In “Which French speech recognition system for assistant robots?” (2022, 6 citations), he provides a systematic comparison of French-language ASR systems, a valuable resource for developers selecting the best tool for robotic assistants. His more recent work, “Deep Reinforcement Learning for Mobile Robots: Overview and Issues” (2024), surveys the challenges and opportunities of using deep RL for autonomous navigation and decision-making. Bouchentouf’s contributions are especially relevant for researchers building interactive robots that must understand novel words, operate in French-speaking contexts, and learn from their environment.