Skip to main content


Social Innovation

R & D

Based on the speech signal processing technology that has been cultivated by Hitachi for many years, Hitachi has developed a speech signal technology for general purpose smart devices instead of special purpose devices.

Hitachi has developed speech recognition technology for smart devices that works well, even in noisy surroundings. By removing background noise, except the speaker’s voice, the technology allows for speech recognition even when the noise level is 70dB*. This is very helpful when using the speech recognition technology in typical urban environments, like airports or shopping malls. In addition, it can automatically detect speech intervals, which means users don’t have to keep pressing a button when they speak. This technology will contribute to the use of multilingual speech translation services at places such as public transportation information centers and retail service counters.

As Japan welcomes more visitors, the demand for multilingual speech translation services has increased. This innovative technology will make communication between tourists and local service counter clerks easier, and it will help lessen the language barrier between locals and visitors. The newly developed speech signal processing technology is capable of reducing noise and enhancing a user's voice, providing accurate automatic recognition of speech intervals. It is also capable of decreasing the input time and enabling simultaneous translation of every phrase making live chat possible and offering advantages over special purpose translation devices. The newly developed technology performs speech processing and translation via the cloud. This means users can easily access the system by installing the dedicated app onto their smart devices.

70dB is comparable to a noisy urban street environment

Content By: Hitachi
Release Date: January 13th 2016