MENU

Self-tuning acoustic-ready chip increases audio focus

Self-tuning acoustic-ready chip increases audio focus

Technology News |
By eeNews Europe



For the subsequent launch of the ZL38040, Microsemi’s first Timberwolf-based product, we caught up with Shahin Sadeghi, Director of Marketing and Applications, Communications.

More specifically aimed at hand-free voice communications, the 55nm CMOS design is powered by Microsemi’s AcuEdge acoustic algorithms, which Sadeghi says make it much easier for customers to deploy enterprise hands-free speakerphones.

AcuEdge’s technology consists of license-free, royalty-free audio IP solutions. It is the compilation of various intellectual properties targeted at processing voice signals in conjunction with noise reduction algorithms, automatic gain control, echo cancellation, psychoacoustic noise reduction, howling detection and rejection. These capabilities are the foundation for enterprise level automatic speech recognition, sound classification and other intelligent decision-making functions based on sound and audio detection.

“We offer the AcuEdge firmware as a complete black box, with a sophisticated GUI-based customizing solution so our customers can self-tune the firmware to their application”, Sadeghi told us.

“Even out in the field, they can field-upgrade the firmware, and the part always self-tunes and self-calibrates, so a system could be moved from one room to another with other reverberating characteristics”, he added. 

The ZL38040 is marketed as an advanced wideband acoustic echo canceller designed to provide leading-edge echo cancellation and noise reduction for HD audio speaker phone and IP phone applications. The company says that before bringing its chips to market, it runs its solutions in an anechoic chamber, against benchmark tests for head acoustics in order to pass and beat every industry standard.

Running the AcuEdge firmware, the ZL38040 can perform beam forming, multi-channel acoustic echo cancellation (AEC), and supports direction of arrival and far field MIC capabilities for zooming and to focus on specific sounds.

“Where we are better than others from a pure DSP perspective is that we have developed hardware accelerators for echo cancellation”, explains Sadeghi, “so with just two mics about 50mm apart, we can open-up or close-up the sound beam we are listening to, with an accuracy of 10º. We detect the direction of arrival and tell the processor where the sound comes from, to focus on that beam.”

While focused on a particular sound beam, the processor could still be listening within a whole conference room, to identify new participants and do the right echo cancellation for their intervention, bringing a form of audio intelligence to the speaker system.
To complete this new product launch, Microsemi offers a comprehensive development package to accelerate customers’ time-to-market via a validated reference design and easy-to-use tools that include MiTuner, an automated AEC tuning kit.

Visit Microsemi at www.microsemi.com/audio-processing

 

Related articles:

CSR : Low noise software delivers 30 dB of noise suppression

Noise cancellation speaker driver ICs produce zero audible hiss

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News

Share:

Linked Articles
10s