Multimodal Visual Concept Learning with Weakly Supervised Techniques
filby 2018-07-31T11:35:06+00:00Multimodal Visual Concept Learning with Weakly Supervised Techniques [...]
Multimodal Visual Concept Learning with Weakly Supervised Techniques [...]
We focus on recovering aspects of vocal tract’s geometry and dynamics from speech, a problem referred to as speech inversion.
We have developed highly adaptive multimodal fusion rules based on uncertainty compensation which are compatible with synchronous and asynchronous multimodal interaction architectures.
We are working on microphone array processing and distant speech recognition, aiming to create hands-free, voice-enabled interfaces for home automation control.