Far-Field Voice Technology for AI Assistant
With the surprise success of Amazon’s Echo and it's voice-controlled assistant Alexa, the smart speaker war is heating up to battle for the hub of home automation. Underneath these similarly-designed intelligent speakers, what's the far-field voice technology involved for Amazon Echo, Google Home, and Apple HomePod? You might heard of the buzz words beamforming and microphone array, but far-field speech recognition is far more complicated than that. In this talk, we'll have a deep dive of the background, system design and architecture on the far-field voice recognition system.
Tao Ma, Ph.D. is the Principal Architect, AI Platform & Research division at JD.com, China’s largest online retailer and the country’s biggest Internet company by revenue. Prior to joining JD.com, Tao was a Senior Speech Scientist of Siri at Apple. As an early member of Siri speech team, between 2012/05 to 2017/10, Tao's work focused on core speech recognition technologies that power Siri, including deep neural networks, statistical language modeling, and weighted finite state transducers. Tao has over 10 years of academic and industrial experience across a broad range of speech and natural language technologies. He received a Ph.D. in Computer Engineering from Mississippi State University. In his early career, Tao worked at Intel Labs, Li Creative Technologies, and Telenav.