Yobe combines patented AI voice-tracking technology and automatic speech recognition capabilities to provide Original Equipment Manufacturers (OEMs) and voice platforms with a robust, real-world interactive voice experience in environments not possible with other voice solutions.
See how we do it
Yobe enables devices to intelligently focus on and track a voice in complex sound environments in much the same way as humans do. The result is better voice isolation, keyword and command recognition over longer distances and in a wider range of environments.
Yobe can interface with any state of the art Automatic Speech Recognition (ASR) platform. Using Yobe as an on the edge “audio front end” enables Natural Language Processing platforms to operate at higher levels of effectiveness in every day noisy environments.
Yobe enables devices to identify speakers through their unique voice signature, enabling biometric identification, accurate voice tracking even in the presence of multiple speakers, and user profile retrieval using only voice.
Yobe’s platform supports the following languages: Arabic, Bulgarian, Catalan, Chinese (Simplified & Traditional), Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, German, Greek, Hebrew, Hindi, Hungarian, Italian, Japanese, Korean, Norwegian, Polish, Portuguese (Brazil), Romanian, Russian, Slovak, Slovenian, Spanish, Swedish, and Turkish
Voice Isolation &
Enhancement
Speech to text
transcription
Language
Translation
Key word
detection
Voice biometrics/
Identification
Yobe’s unique, patented technology filters out noise close to the device to listenYobe’s ability to track and separate voices for improved accuracy for Wake-words and commands (Speech to Text) in noisy environments by 78.9% and 76.2% respectively. In addition, our solution extends the Far-field effective range by up to 20X in near-field noise far-field voice situations.
* Far-field voice in noisy environments
Highly portable “AI on the edge” solution
Improved voice detectability and enhancement in high noisy environments
Voice biometrics for enhanced security and profile generation
Speech transcription
Language translation
Heightened keyword spotting
Yobe is an AI software company based in Boston that has brought AI down to the edge, introducing intelligent behavior much earlier in the process of voice recognition. For developers of voice-enabled software and devices seeking listening performance in noisy and complex sound environments, Yobe, Inc. provides sound enhancement for voice that delivers human-like responsiveness through AI and studio-level sound processing.
President & CEO (Co-Founder)
Chief Scientist/Technology Advisor(Co-Founder)
Ken is a serial entrepreneur with nearly 20 years of strategy and corporate business management experience in the technology, marketing and finance industries. In addition to his role as the lead technology evangelist, Ken manages the company’s product vision, corporate development, team building, brand management, and investor relationships.
Prior to Yobe, Ken was a financial services professional who worked on critical projects for hedge funds, venture capital and private equity firms focused on Pre-IPO investments in the technology and real-estate arenas. Ken’s formal entrepreneurial experience started directly after University with the creation of The Tampa Marketing Group. TMG was a multi-state marketing firm, focused on product development, market research and brand strategy. Other start-up experience includes: Founding Member of Co-Mune Inc (a community-focused connectivity and mobile commerce platform), and Managing Partner Sutton Willis and More (a boutique strategy and capital advisory firm).
Ken, a Hartford CT native, attended the University of Connecticut and has served as a proud member of the US Armed Forces (US Army Ranger).
An internationally renowned researcher, engineer, and educator, Nawab (PhD, MIT ’82) conducts research at the intersection of signal processing and artificial intelligence for applications in speech, audition, and neuroscience. A Professor of Electrical & Computer Engineering and Biomedical Engineering at Boston University, he is an elected fellow of AIMBE (Class of 2006) that represents the top 2% of medical and biological engineering researchers. He has published over 100 research articles and co-authored the seminal book on “Symbolic and Knowledge Based Signal Processing” (Prentice Hall, 1992).
Dr. Nawab, is supported by a team of three commercial algorithm engineers with experience in conceptualization, implementation, and evaluation of signal computing with an emphasis on applied artificial intelligence.
Yobe can work with OEMs to fit your business model. We offer a range of licensing and subscription options, including perpetual unit-based licensing fees, subscriptions, and SaaS options.
Contact us to learn more about Yobe, schedule a personal consultation and learn how our cognitive software can improve the experience for your voice-enabled service.
