Audio Weaver, which s already being used by a number of customers, is a graphical configuration tool that enables the simple and rapid development of audio/voice DSP software. It eliminates the need for native coding and kernel optimisation and can be used to easily design, debug, and tune audio processing flows in real-time, cutting product development time by up to 90%.
TalkTo is a high performance far-field voice front-end that leverages Audio Weaver for ease of integration and tuning. When used in conjunction with CEVA’s audio and voice DSPs, product makers will be able to fully exploit the processing performance of the DSP engine and create unique audio and voice features to differentiate their products.
The CEVA-X2 and CEVA-BX family of DSPs are intended any power-constrained audio/voice use case and are widely adopted to date in SoCs for smartphones, smart speakers, voice assistants, action cameras, and more.
“Embedded software development for audio and voice applications tends to be complex, time consuming and costly,” said Dr. Paul Beckmann, Founder and CTO of DSP Concepts. “Our Audio Weaver Design Tool significantly simplifies this process for product designers and we’re pleased to offer these capabilities now to CEVA’s broad customer base integrating DSP-based audio and voice functionality into their SoC designs.”
“Our collaboration with DSP Concepts brings the benefits of their Audio Weaver tool to our customers, providing an impressive design environment to build out even the most complex audio and voice use cases, in minimal time,” said Moshe Sheier, Vice President of Marketing at CEVA. “We’re already receiving excellent feedback from our joint customers and look forward to expanding our partnership as audio and voice adoption in the everyday products that surround us continue to grow.”
CEVA’s high performance audio and voice DSPs are optimised for intensive sound processing applications and have been specifically designed to tackle multi-microphone speech processing use-cases, high quality audio playback and post-processing, and on-device sound neural network implementations. These DSPs are supported by comprehensive software offerings for speech recognition and voice processing that have been developed and optimised in-house at CEVA.
