STMicroelectronics and Sensory Collaborate to Enable Mass Adoption of Embedded Voice Control

Article by: STMicroelectronics

STMicroelectronics and Sensory are collaborating to enable the STM32 microcontroller user community to develop and prototype intuitive voice user interfaces for a wide range of smart embedded products.

STMicroelectronics and Sensory Inc., an in-vehicle speech recognition technology provider and an ST Authorized Partner, are collaborating to enable the STM32 microcontroller (MCU) user community to develop and prototype intuitive voice user interfaces for a wide range of smart embedded products.

Joint efforts combine ST’s STM32 hardware and software with Sensory’s voice control technologies, including the new VoiceHub online portal that supports the seamless creation of embedded speech recognition models using custom wake words , sets of voice control commands and great natural language grammars in nearly twenty languages ​​and dialects.

The solution is based on an STM32Cube software expansion package and runs on a high-performance STM32H7 microcontroller, taking advantage of its architecture, internal flash memory, SRAM and high processor speed. This combination plays a key role in increasing voice command accuracy and reducing command recognition time. Hosting the voice application and voice models in the generous on-chip memory of high-performance STM32 microcontrollers further improves system integration and ease of use, while lowering the cost of ownership.

“This collaboration will kick-start the development of embedded voice user interfaces, adding frictionless command control and a personalized wake word to any device, from wearables to smart home devices,” said Ricardo De Sa Earp. , Executive Vice President, Vice President of General Purpose Microcontrollers Subgroup, STMicroelectronics. “The unique combination of ST and Sensory technologies will allow the STM32 user community to deploy ‘Voice AI on edge’ without any programming, data science or machine learning expertise, for free in prototypes and with licensing terms. favorable in production. ”

“Sensory designed our VoiceHub so that developers can quickly and painlessly create custom speech recognition models. However, after building a custom model, integrating the model on hardware and moving to license terms were the next hurdles to overcome,” said Todd Mozer, CEO of Sensory. “This world-class collaboration with ST creates a complete set of software, hardware and licenses for voice recognition integrated into the STM32 family and simplifies the addition of voice user interfaces.”

Comments are closed.