Description
Product Type:
Voice generator
Prototype
Description:
The company is developing an assistive product that uses machine learning and sensor technology to give tumor patients back a natural-sounding voice after losing their larynx.
Radar sensor technology, a non-invasive sensor technology, can detect and characterize silent speech movements. The latest deep neural networks are powerful enough to convert these signals into audible speech in real time.
The technology is being developed in collaboration with the Chair of Speech Technology and Cognitive Systems and the Chair of High Frequency Technology at Dresden University of Technology.
The joint project SEMECO-B3 Multi-sensory non-invasive voice prosthetics using AI (MUSIK) is funded by the Free State of Saxony and co-financed by the European Union.
Manufacturer contact data
