Vocal Synthesis Jay-Z AI voice impersonations
An pseudonymous music creator named 'Voice Synthesis' used AI to generate deepfaked tracks of rapper Jay-Z's reciting Shakepeare's 'To be, or not to be' and Billy Joel's Don't Start the Fire. Voice Synthesis’ videos created the deepfake videos by feeding Google’s Tacotron 2 text-to-speech model with Jay-Z's songs and lyrics, and having the synthetic voice read pre-written text. Other videos created by Voice Synthesis include Tucker Carlson reading the Unabomber Manifesto and Bill Clinton reciting 'Baby Got Back '. Jay-Z's agency entertainment Roc Nation LLC claimed copyright infringment and argued 'This content unlawfully uses an AI to impersonate our client’s voice.' YouTube took down the videos, but later reinstated them on the basis that the DCMA request was 'incomplete'. The videos remained on decentralised, open source platform LBRY. Input noted that Voice Synthesis 'transformed Jay-Z’s discography in a humorous way for no commercial benefit and clearly labels all videos as speech synthesis'. System 🤖 Tacotron 2 Operator: Vocal Synthesis; Google/YouTube; LBRY Developer: Vocal Synthesis Country: USA Sector: Media/entertainment/sports/arts Purpose: Entertain Technology: Deepfake - image; Generative adversarial network (GAN); Neural network; Deep learning; Machine learning Issue: Mis/disinformation; Ethics/values
- Date it happened
- 2020-04-01
- Organisation involved
- Vocal Synthesis; YouTube; LBRY
- Product, system or model
- Tacotron 2
This incident was imported from AIAAIC and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.
This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.