Developers: | Promobot |
Date of the premiere of the system: | November, 2020 |
Technology: | Robotics |
2020: Announcement of a system
On November 20, 2020 Promobot announced creation of a microphone array which in the company is called "ears" of the robot. The new system consisting of eight microphones and a program algorithm will recognize the speech in noisy places, including in the airport, MFC and shopping centers.
Thanks to technology robots can identify noise at distance to two meters from it, detection range — 360 degrees, as at the person. The algorithm provides capture of analog data from microphones, their conversion, echo suppression and data transmission.
The program "separates" the human speech from noise — the inner ear in a human body in the same way works. Then the algorithm distinguishes phrases thanks to what the robot enters dialog. Unlike voice assistants who react only to the accurate speech of the person Promobot reacts to all noise — and already "in it" looks for a human voice, and both children's, and the adult, says the producer.
Unlike other systems, Promobot speech recognition practically does not depend on distance: in a year the recognition percent at distance of one meter from the robot grew three times - from 25% to 75%. It is the highest rate in the world for service robots, and it, by words the developer, as much as possible brings closer work of a microphone array to a human ear.
The feature of speech recognition in service robotics is that our devices work in places of the increased accumulation of people. Microphones in smartphones, earphones and smart columns are personal devices, they are intended for work only with the owner, is frequent — only in house conditions — the development director of Promobot Oleg Kivokurtsev noted. |
The beginning of sales of innovation is planned for January, 2021.[1]