3 ms·Audio Deepfake: Text-to-Speech trained on less than 15 minutes of audio5 points by aracena 7y ago