The i.MX RT600 is a cross-over processor family optimized for 32-bit immersive audio playback and voice user interface applications combining a high-performance Cadence Tensilica Hi-Fi 4 audio DSP core with a next-generation Cortex-M33 core. The i.MX RT685 family provides up to 4.5MB of on-chip SRAM and several high-bandwidth interfaces to access off-chip flash. In addition, the family offers a rich set of peripherals, advanced security capabilities and extremely low-power consumption with reduced power modes and fast wake-up times. The processor is now in pre-production stage.
The Alango Voice Enhancement Package (VEP) is a suite of real-time software DSP technologies designed for improving speech recognition performance in voice-controlled multimedia devices. VEP is a front-end solution that performs audio signal pre-processing prior to the Automatic Speech Recognition (ASR) and Key Word Recognition (KWR) engines. VEP is an ASR-engine agnostic signal processing package that can be used in conjunction with any ASR engine. The technologies included in VEP are universal and do not interfere with signal processing performed as part of modern ASR’s.
The development solution includes multi-microphone acoustic beamforming, stereo echo cancellation, noise reduction, automatic gain control, and direction of arrival finding. The package is fully configurable supporting various microphone configurations and direction finding strategies. VEP has been optimized for a variety of DSP and MCU cores including ARM Cortex, Cadence Tensilica HiFi, and Synopsys ARC.
Alango cooperated with NXP and Cadence to demonstrate an early stage porting of Alango Voice Enhancement Package (VEP) at CDNLive in Taiwan on August 2019. This porting fully complies with NXP Xtensa Audio Framework (XAF). The Cadence Tensilica HiFi DSP audio framework provides a standard operating system (OS) agnostic software framework to interface audio, voice, and sensor functions running on the HiFi DSP to higher level applications.