It can be challenging for people with even a slight hearing loss to hear the dialogue in movies and TV shows over the background music and special effects. An AI developed by researchers at Mitsubishi Electric Research Laboratories, Merl, in the United States may soon reduce those problems, reports Unite.AI.
The researchers have trained the AI to recognize speech, music, and special effects. It then separates them into three different channels. Once done, the viewer can specify the volume for each sound category by using an app.
Another area of use may be sound mixing. The AI makes it possible for anyone who wants to mix the sound in their own Youtube videos, for example, to get a good mix without knowing anything about sound mixing.
There are already other systems to be able to change the sound mix. Swedish public service TV network SVT has, for example, tested a different method. The method developed by Merl researchers has the advantage that it should not depend on the original soundtrack being suitable for changing the mixing.
The researchers hope that the technology in the future will be built into smart TVs.
Image: Pixabay / StockSnap
