CPC G10L 21/0272 (2013.01) [G10L 15/20 (2013.01); G10L 25/48 (2013.01)] | 10 Claims |
1. An acoustic scene conversion method, comprising:
receiving sound signals including user's voice sounds and scenic sounds;
receiving positional data and speed of the user from a positioning device based on satellite positioning system;
selecting an AI model from multiple candidate AI models according to gender, age, the positional data and the speed of the user, wherein the gender and the age of the user are set up by the user prior to the selecting;
processing the sound signals according to the AI model in order to generate enhanced voice sounds without scenic sounds;
mixing the enhanced voice sounds without scenic sounds with new scenic sounds in order to generate converted sound signals;
transmitting the converted sound signals to a remote apparatus for playing via a network in real time; and
having the remote apparatus play the converted sound signals.
|