Spatial Speech Translation consists of two AI models, the first of which divides the space surrounding the person wearing the headphones into small regions and uses a neural network to search for potential speakers and pinpoint their direction.
…
A new AI translation system for headphones clones multiple voices simultaneously
