Different systems of artificial intelligence already know how to do such things that it’s time to think about the fact that the plot of the film franchise Terminator is not such a fantasy. Moreover, every day AI learns new and new techniques. For example, recently, Google specialists trained their artificial mind to identify the voices of individuals in the crowd noise with astonishing accuracy.
To begin with, Google engineers created a technology based on machine learning, which reveals the voices of individuals through the “party effect”. This is what we all are able to do: communicating with a person surrounded by extraneous noise, we can understand what he is talking about, analyzing not only speech but also facial expressions, filtering out “disturbing” sounds. AI Google works in a similar way: by analyzing facial muscles and all the noise, it highlights the voices of different people and creates separate audio tracks from them. Then Google learned how to recognize the artificial intelligence of people’s faces, isolate their voices and suppress extraneous noise.
Thus, the system can allocate from the crowd of noisy people the necessary personnel and make recognition of his voice in real time. This became possible due to the fact that Google engineers “trained” AI on more than 100 000 video lectures and conversational videos from YouTube with background noise. The new development, according to Google, can be used, for example, to improve the quality of video chat, so that during group communication the user could focus on one interlocutor. The result of the work of the algorithm Google you can watch the video, available below.