FREAK: Frequency-modulated High-fidelity and Real-time Audio-driven Talking Portrait Synthesis

Re-assign community