Leveraging the patented Speech-to-Meaning and Deep Meaning Understanding technologies from SoundHound Houndify Voice AI platform, running on NVidia Drive IX, the solution enables real-time responses to voice queries in vehicles, even without Internet connectivity. This is achieved with high speed and accuracy through a hybrid speech recognition system that processes voice requests both in the cloud and locally on the embedded system (for when an Internet connection is not available) to return fast responses. The embedded system also enables drivers to control their car’s functions when a connection to the cloud is unavailable including the car’s climate control, window controls, radio, navigation, and more.
NVidia Drive AGX integrates the high-performance, energy-efficient compute of the NVidia Xavier system-on-a-chip (SoC) and full stack AV software to monitor surroundings and the driver, localize to an HD map, and plan a safe path forward. Within Drive software, NVidia Drive IX is a framework for the full cockpit experience. It combines the system, tools, and algorithms to enhance the driver’s situational awareness, assist in driving functions and provide intelligent interactions between the vehicle and its occupants. SoundHound understood this as an ideal platform for integrating the voice technology that Houndify can provide, enabling the vehicle to seamlessly respond to human voice commands.
“The NVidia Drive platform has enabled us to create an embedded solution for interacting with cars using voice and natural language,” says Keyvan Mohajer, Founder and CEO, SoundHound Inc. “By using NVidia GPUs for deep learning training, and the Drive IX platform for embedded computation using the GPU inside the Xavier SoC, we are able to scale to large vocabulary in natural language with the Houndify platform, maintaining speed and accuracy, even without a cloud connection.”
“Low-latency speech recognition is an important aspect of intelligent experiences in the vehicle,” adds Danny Shapiro, senior director of automotive at NVidia. “SoundHound’s innovative solution on our open Drive IX platform will allow carmakers to offer systems that have an enormous vocabulary, understand a wide range of topics, and respond conversationally.”
With Houndify, drivers can also interact with hundreds of domains — programs that provide users with relevant information or actions related to their queries. These include: navigation, weather, stock prices, sports scores, flight status, local business searches, and hotel searches with complex criteria, among others.
SoundHound’s Houndify technology is already being applied by car manufacturers including Mercedes-Benz, Groupe PSA, Hyundai, Honda, and others. SoundHound had been in stealth mode with its voice technology and natural language understanding research and development for a decade. The company quietly built the technology stack needed to create a complete solution that enables companies to deploy a customized, branded voice experience. The Houndify platform enables developers to use that proprietary technology in their own products. SoundHound has two consumer apps powered by Houndify: Hound, the voice assistant app, and SoundHound, the popular music discovery and lyrics app, with over 310 million unique downloads globally.
Developers interested in exploring the Houndify platform can visit www.houndify.com to sign up for a free trial and learn more. Additional information on NVidia Automotive solutions can be found by visiting www.nvidia.com/drive.