Areas Bertrand Delvaux is Knowledgeable in:
Acoustics, Physics Modelling, Algorithms
Bertrand Delvaux's Problem Solving Experience:
- I imagined, designed and conceived entirely a Python script (MapMyFile, © May 2016) whose algorithm automatically detects and tunes customized features in car’s Engine Control Unit (ECU) for better performance (power, torque, speed, consumption, ...). T he first software in the world that automates detection and tuning , it reduces tuning time from 2h (manual) to 2.5 s. So far, 70K+ car files (models and versions from VAG group and BMW) can already be generated that way.
- I investigated solo voice acoustics. Specifically, I scanned singers with MRI, reconstructed
their vocal tract in 3D to simulate numerically their impulse response. I 3D-printed their vocal tract and measured their acoustic impulse response in an anechoic chamber. These were the first full vocal tracts printed in the world . I found the spectral impact of morphological features of the hypopharynx on a singer’s voice. I created a new metrics that determines the voice classification and spectral gaussian curve depending on morphological dimensions.
That technic was later used to reconstruct a 3000 year old mummy's voice (see supplementary information on https://www.nature.com/articles/s41598-019-56316-y)