Malaspina Labs announces ultra-low power speech recognition and biometric authentication solutions for STMicroelectronics STM32 microcontrollers
June 1, 2015 —
Malaspina Labs announces the availability of its VoiceBoost™ suite of speech processing solutions for the STM32 family of microcontrollers from STMicroelectronics. The VoiceBoost suite of speech processing solutions includes noise reduction, ultra-low power voice wakeup, biometric speaker verification, and embedded speech recognition products.
The solutions will be demonstrated at Computex Taipei, June 2-6 2015.
The VoiceBoost Speaker Independent Voice Activation solution allows users to wakeup and interact with STM32-based devices using a spoken phrase, without requiring typing, gestures or button presses. The solution executes continuously in a low power state, drawing an incremental current of less than 0.17mA at 1.8v on average in moderately noisy environments. The complete system including microcontroller, speech solution, microphone and all integrated peripherals consumes less than 1mA on average in moderately noisy environments. The solution also allows “phrase spotting” with or without preceding wakeup phrases, to control device functions by voice commands.
Malaspina’s VoiceBoost Biometric Speaker Verification solution allows STM32-based devices to implement biometric voice authentication, so that the device will respond only when an enrolled user’s private passphrase is spoken by the user’s own voice. The Biometric Speaker Verification solution allows manufacturers to add device locking, parental controls and multi-user capabilities to smart consumer and IoT devices.
Like the Speaker Independent Voice Activation solution, the VoiceBoost Biometric Speaker Verification solution is effective in high background noise with users speaking at arm’s length and beyond, and can execute in an ultra-low power “always listening” mode. The Biometric Speaker Verification solution adds just 0.08mA at 1.8v incremental current consumption on average in moderately noisy conditions. The complete system, including microcontroller, speech solution, microphone and all integrated peripherals, consumes less than 0.9mA on average in moderately noisy environments. In addition to authenticating users for access to a device, the Biometric Speaker Verification solution can be used in conjunction with the Speaker Independent Voice Activation solution to allow limited device functionality for all users, while providing personalized functions and information to identified users. All VoiceBoost solutions execute entirely locally on the STM32 device without requiring any Internet connectivity or cloud-based services.
“The Smart Wearables market is growing rapidly and for many of these small and low-power devices that can’t use touch-screen displays because of size and cost, voice commands can do the job,” said Michel Buffa, General Manager Microcontroller Division, STMicroelectronics. “Malaspina’s VoiceBoost technology is an excellent solution for that use case and it is a great match for the STM32 microcontrollers. It provides robust performance in noisy environments with minimal power consumption.“
Dean Neumann, CEO of Malaspina Labs said that “STMicroelectronics' large range of STM32 microcontrollers are a great solution for wearables, smart consumer devices, IoT and industrial devices wanting to provide voice control or multi-sensor processing capabilities at ultra-low power. We are delighted to be able to offer our full suite of speech solutions highlighting the performance and range of the STM32 microcontrollers”.
About Malaspina Labs
Malaspina Labs develops speech interface and enhancement software for ultra-low power devices which improves person-to-person and person-to-device voice communication in highnoise environments. Malaspina Labs' VoiceBoost™ suite of solutions provides speech isolation and voice interface technology that is effective at handset, speakerphone and arm's length distances and beyond. VoiceBoost solutions operate deterministically at bone-conduction, telecom, and voice interface latencies within the constraints of milliwatt-class processors designed for mobile, embedded, medical and consumer devices.
For more information about Malaspina Labs, visit