About this role
Develop and implement wake word detectors, optimizing a local speech recognition engine, and creating an event detector to listen for key sounds such as glass cracking. Preferred candidate has detailed, hands-on knowledge of modern ASR techniques, experience building speech engines for computationally constrained local environments without any server support, experience building models for domain specific ASR, the ability to write shippable, production grade software, knowledge of OpenCL or other GPU compatible parallel computing frameworks.
All your information will be kept confidential according to EEO guidelines.