The Role of Speech Acoustics to Detect Deep Fake Voices

Supervisor: Dr. Elisa Pellegrino

A corpus of synthetic utterances (voice converted utterance) created such that they have the prosody of the source speaker and the timbre of the target speaker is available for testing the hypothesis that acoustic differences between source and target speakers can be used to distinguish in production and perception natural voices from their synthetic copies

• Praat
• R
• Online experiment builder