SoundCTM: Uniting Score-based and Consistency Models for Text-to-Sound Generation