DSP Concepts Introduces Audio Framework for Cadence Tensilica HiFi 5 DSP

DSP Concepts, an audio platform provider, today announced early access availability of its Audio Weaver library for the Cadence Tensilica HiFi 5 digital signal processor, which is scheduled for general release by the end of June.

The Tensilica HiFi 5 DSP includes special design elements for enabling rich voice-control user interfaces and artificial intelligence (AI)-based speech recognition.

Combined with optimized Audio Weaver libraries, Tensilica HiFi 5 DSP will have access to the full suite of DSP Concepts' proprietary and third-party intellectual property. This includes DSP Concepts' audio front-end (AFE) TalkTo, which provides the Adaptive Interference Canceller and multichannel echo cancellation with advanced beamforming using eight or more microphones. Additionally, the new machine learning (ML) instruction sets in the Tensilica HiFi 5 DSP enhance full duplex communication and playback processing.

"The exponential growth in adoption of voice interactions for a vast array of devices has spurred the need for increasing signal processing and neural-network performance," said Yipeng Liu, product marketing group director for Tensilica audio/voice IP at Cadence, in a statement. "To unlock the full capabilities of voice, we have collaborated with DSP Concepts to enable their high-performing algorithms and Audio Weaver on the Tensilica HiFi 5 DSP. Optimized on the Tensilica HiFi 5 DSP, Audio Weaver smoothly runs these algorithms, providing clear voice communication in the most challenging environments. The ability to extend talk time with best-in-class energy consumption across advanced AI use cases, including scene analysis and natural language processing, sets the Tensilica HiFi 5 DSP apart from other DSPs."

At the center of these innovations are DSP Concepts' smart audio and voice solutions that include far-field voice UI for smart assistants, TVs, soundbars, and set-top-boxes. These solutions are also optimized for voice communication and home and office video conferencing with adaptive interference canceller and acoustic echo cancellation for up to seven channels.

"The drive toward higher performance, no matter what the application, has never been stronger, particularly for headsets, wearables, and battery-powered edge devices," said Michael Maia, vice president of marketing at DSP Concepts, in a statement. "Powerful processing cores like the Cadence Tensilica HiFi 5 DSP drive innovation forward rapidly, and our close collaboration with Cadence allows designers and devices adopting this solution to not only keep pace with these accelerating market demands, but in turn, set that pace with support for new, innovative use cases. There is no other solution in the market with this level of performance specifically designed for the future possibilities of AI and machine learning."

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

Cadence Launches On-Device Tensilica AI Platform

Cadence's new Tensilica AI engine boosts performance, and AI accelerators provide a turnkey solution for consumer, mobile, automotive, and industrial AI SoC designs.