CEVA Inc. has revealed that China-based iFLYTEK's speech recognition software is available for CEVA's audio/voice DSPs.

Speech recognition is gaining popularity, just behind sound processing advances and artificial intelligence, as the human machine interface (HMI) for consumer electronics, mobile, smart home and wearable devices, as well as surveillance, automotive and IoT.

CEVA and iFLYTEK collaborated to optimise iFLYTEK's neural network based speech recognition, noise reduction and echo-cancellation algorithms for CEVA's advanced audio/voice DSPs. The result is a powerful and highly accurate, on-device voice processing solution, capable of enabling multiple mics voice activation without requiring cloud access, the companies said.

CEVA's family of audio/voice/speech DSPs provide a range of solutions, from ultra-low-power always-listening smart devices, through versatile wireless audio in mobile and hearables, and up to high-resolution audio processing in home entertainment and automotive. These audio/voice DSPs are ideal for developing voice-enabled solutions where power consumption and cost are critical.

The integrated solution is available for customers and is already in production for an ultra-low power voice processor targeting high volume consumer electronics.