US10108606B2 - Automatic Interpretation System and Method for Generating Synthetic Sound Having Characteristics Similar to Those of Original Speaker's Voice