Ismail Rasim Ulgen

Enhancing speech synthesis for more natural and expressive voice generation.

SelectTTS simplifies speech generation for unseen speakers with effective frame selection.

New masking method improves voice conversion by separating speaker identity from phonetics.