Smart headphones have been translated into several speakers at once
Researchers at the American University of Washington developed the smart headphone system, which can translate many people at the same time, while at the same time, with three -dimensional accuracy of each speaker’s voice and the direction of his voice, with three -dimensional accuracy.
Researchers have pointed out that this system, known as “territorial translation”, can make a qualitative change in the simultaneous translation field. The results were published on Friday in the American Computing Association (ACM).
Despite the emergence of good methods in the recent translation sector, none of them provide an effective solution in public places. For example, new meta glasses only operate when a speaker and automatic voice translation is displayed after completing the speaker speech. But the team designed for the new system has found a solution that translated the words of many speakers during the foresight while maintaining the direction and characteristics of each voice.
This system relies on the sound of microphones -relying on the headphones, and to separate the throats of the speakers and follow them inside the space, and then translate their words on the system of artificial intelligence, and then translate their words in their original voice between 2 and 4 seconds.
The system is classified by three major innovations, the first of which around the 360 -Degree Ocean, any closed or
In partnership with 29 volunteers, the system has been tested in 10 different internal and external environments, such as the streets, cafes, museums, public places and its effect in these realistic surroundings, despite the background noise. The experiment focuses on 3 languages, Spanish, German and French, and this system has shown good performance in translation from these languages to English and on the contrary.
The results showed that most users prefer a new translation system on traditional systems; This is because he has provided a natural auditory experience that mimics high accuracy and reality in translation. The rule of the regime demonstrates its high efficiency in the followers of the speakers and continues to separate the direction of each sound and its expressive characteristics, even determining their number when they move.
When opening new limits for tourism, education and cultural interactions, preserving their original voices, preserving their original voices, preserving their original voices, when preserving their original voices, in congestion environment, this system allows users to understand what many people are saying at the same time. The team hopes that this innovation can expand about 100 languages that can break linguistic obstacles amidst the cultures that can expand about 100 languages so that the user can understand what is happening in its language.
