Mitsubishi Electric : Separates Simultaneous Speech of Multiple Unknown Speakers Recorded with One Microphone
May 23, 2017 at 11:46 pm EDT
Share
In the case of two simultaneous speakers, accuracy exceeded 90 percent, sufficient for commercial applications, compared with 51 percent accuracy using conventional technology. The new technology is able to discern between combinations of several spoken languages and gender. The above results are based on ideal recording conditions, including low ambient noise and speakers talking at roughly similar volume. The Deep Clustering technology uses Mitsubishi Electric's proprietary deep-learning method to learn how to encode signal components of the original speech data of multiple people so that signal components belonging to each individual speaker can be easily distinguished by their encodings. To accomplish this, the encodings are optimized such that different signal components belonging to the same speaker have similar encodings, and those belonging to different speakers have dissimilar encodings. The learned encoding transformation is applied to the input speech, and the encodings of the signal components of each speaker are identified using a clustering algorithm, which processes data points into groups depending on their similarities. Each person's speech is then reconstructed by resynthesizing their separated speech components.
Mitsubishi Electric Corporation published this content on 24 May 2017 and is solely responsible for the information contained herein. Distributed by Public, unedited and unaltered, on 24 May 2017 03:45:15 UTC.
Original documenthttp://www.mitsubishielectric.com/news/2017/0524-e.html
Public permalinkhttp://www.publicnow.com/view/F813E051290F4504C763485DB2B8ED657A476547
Mitsubishi Electric Corporation is one of the world's leading manufacturers of electric and electronic equipment. Net sales (including intragroup) break down by family of products and by services as follows:
- industrial automation equipment (28.6%): programmable controllers, motor operators, counters, industrial robots, clutches, etc.;
- electric equipment (24.2%): generators, motors, transformers, power circuit breakers, testing systems, etc.;
- household appliances (22.4%): televisions, video recorders, air conditioners, refrigerators, lamps, vacuum cleaners, etc.;
- information and communication equipment (6.9%): mobile phones, satellites, aerials, medical and audiovisual equipment, servers, computers, etc.;
- electronic components and devices (4.7%): memory, logic integrated circuits, monitors, plasma screens, printed circuit cards, etc.;
- other (13.2%): financial services, real estate management, engineering and maintenance services, etc.
Net sales break down geographically as follows: Japan (52.1%), Asia (24.9%), Europe (11.1%), North America (10.3%) and other (1.6%).