R&D Stories: Speaker ID Technology

As part of its focus on voice convergence, in its April 2021 edition, audioXpress featured a contributed article by the founder and CEO of Fluent.ai, describing how increasing consumer demand for personalized experiences has sparked the creation of the company’s speaker ID technology, and exploring the challenges that this type of technology still faces. A valuable perspective from a company positioned right at the edge of voice recognition using AI and edge processing. This article was originally published in audioXpress, April 2021.

It's been precisely a year since the global COVID-19 pandemic was declared and we all had to refrain from our normal routines and shelter in place, being restricted to working from home until now. At least for the audioXpress team. A year past, and we are glad to be able to signal that during that year we've kept our publications and even grown our online audiences and achieved extremely positive results. Mainly, because we continued to offer the best content for anyone interested in following the audio industry, audio product development, audio innovations, and DIY audio.

The April 2021 issue of audioXpress is now available in print and online and it addresses a convergence of technologies in two of the most exciting product segments at the moment: True Wireless Earbuds and Voice & Communication. George Ntanavaras explores vibration measurements using piezoelectric sensors, Jens Tybo Jensen describes a highly efficient hybrid bass amplification design, and Richard Honeycutt discusses how horns have evolved to address sound directionality. And much more top-level audio electronics from top experts in the field. Get Your Copy Now!

CEVA and Fluent.ai, a leading provider of on-device, multilingual speech understanding solutions, announced a partnership to offer ultra-low power speech-to-intent solutions for intelligent edge devices. Fluent.ai’s suite of speech-to-intent technologies has been ported and optimized for CEVA’s low power audio and sensor hub DSPs, providing a high performance, robust solution for OEMs and ODMs looking to integrate intelligent voice activation and control into their wearables, consumer devices and IoT products.

Fluent.ai, the speech recognition provider offering fully embeddable speech recognition solutions in any language, announced the release of its ultra low-power voice control solution for embedded platforms such as Cortex M4F processors. This breakthrough new solution delivers, using a single engine, multiple wakewords in parallel and up to 50 multilingual commands with variable phrasing on a footprint equivalent to 50 MHz CPU, 128KB of RAM, and 550 KB of flash storage - a quantum leap over currently available software solutions.

Fluent.ai Inc., the company that claims to have created "the world’s first acoustic-only multi-lingual voice interface" is now offering an intuitive and flexible Platform-as-a-Service, FaaS, for developers. According to the company, the platform allows implementations of the company's artificial-intelligence solution to "quickly customize reliable and robust voice capabilities for consumer devices and cloud service applications."