CEVA Announces DSP and Voice Neural Networks Integration for Power-Efficient Edge Processors

CEVA announced that its CEVA-BX DSP cores and WhisPro speech recognition software now also support TensorFlow Lite for Microcontrollers, a production ready, cross-platform framework for deploying tiny machine learning on power-efficient processors in edge devices. WhisPro is CEVA's speech recognition software for voice wake words and custom command models. TensorFlow Lite for Microcontrollers is a solution from Google for implementing machine learning at the edge, already optimized and available for CEVA-BX DSP cores.

The market for virtual assistants (VAs) continues to illustrate strong positive momentum, according to a new report from Futuresource Consulting. Shipments of products supporting virtual assistants rose 25% year-on-year to 1.1 billion units in 2019, with the market on course to exceed 2.5 billion shipments by 2023. Apple leads the market in terms of products sold with support for its Siri voice assistant.

Bragi, the company which can rightfully claim to be the creators of the world’s first Hearable, is unveiling the next-generation of its technology suite and is to introduce it at MWC 2019. As previously announced, Bragi is now developing ultra-efficient artificial intelligence and edge computing technologies for ultra-low power devices that intends to make available to other manufacturers. At MWC 2019, the German company will disclose details of the new platform as a reference design for professional Hearables.

For developers working on industrial robots, drones and all sorts of connected or smart home devices, the new Intel Neural Compute Stick (NCS) 2 offers what's needed to prototype faster and smarter, runs on a standard USB 3.0 port and requires no additional hardware. The NCS 2 was introduced at Intel's first artificial intelligence (AI) developer conference in Beijing (Nov. 14-15, 2018), an event were the company promoted its vision for building smarter AI algorithms and intelligent devices at the network edge.

Cadence Design Systems announced the Cadence Tensilica HiFi 5 DSP for audio and voice, the company's first IP core optimized for high-performance far-field processing and artificial intelligence (AI)-based speech recognition processing. This fifth-generation HiFi DSP offers 2X audio processing and 4X neural network (NN) processing improvements versus the HiFi 4 DSP, making it ideal for voice-controlled user interfaces in digital home assistants and automotive infotainment.

Cadence Design Systems announced the Cadence Tensilica DNA 100 Processor IP, the first deep neural-network accelerator (DNA) AI processor IP to deliver both high performance and power efficiency across a full range of compute from 0.5 TeraMAC (TMAC) to 100s of TMACs. As a result, the DNA 100 processor is well suited for on-device neural network inference applications spanning autonomous vehicles (AVs), ADAS, surveillance, robotics, drones, augmented reality (AR) /virtual reality (VR), smartphones, smart home and connected devices.