A Subject-Specific Acoustic Model of the Upper Airway for Snoring Sounds Generation