Deepfake Speech Detection Engineer

IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires is seeking a talented and motivated Postdoctoral Researcher in Computer Science to join our esteemed Expression research team.

The Expression team at IRISA is at the forefront of AI innovation, specializing in the study and generation of human language across various modalities: text, speech, and sign. We are actively involved in a project dedicated to developing and evaluating cutting-edge deepfake speech detection systems.

As part of this crucial initiative, we require a skilled researcher to contribute to the massive generation of synthesized speech for deepfake detection purposes.

• Develop and implement advanced speech synthesis systems encompassing a wide range of technologies, including voice cloning and voice conversion.
• Prepare comprehensive datasets for diverse languages.
• Establish a robust framework for Text-to-Speech (TTS) synthesis.
• Implement various TTS systems tailored to different languages.
• Actively contribute to the development of innovative deepfake detection systems.

We are seeking a highly motivated individual with a strong academic background and proven expertise in the following areas:

• Software engineering proficiency in C++, Python.
• In-depth knowledge of machine learning methods and tools (TensorFlow, PyTorch, Keras).
• Expertise in Automatic Speech and Language Processing.
• Familiarity with CI/CD principles.

A PhD in Computer Science or Signal Processing is required for this position.