User:.hecko/Uberduck

Uberduck is an online text-to-speech and video creation service. It offers free access to community-contributed voice clones of various celebrities and fictional characters.^{[2] [3] [4]} The unauthorized voice clones are meant for entertainment purposes, and the terms prohibit commercial or defamatory usage.^[5] The website includes a video editor with direct integration to the text-to-speech engine and a gallery of user-submitted images. An API can also be used to programmatically generate videos with custom voice lines.^{[u 1] [6]}

Uberduck was created in 2020, initially as an audio chatbot with custom text-to-speech voices. However, the creator found that "speech synthesis was the most exciting part of the project" and started focusing on it instead.^{[u 2]} It became popular on TikTok in May 2021 as users began using it in protest of the app's built-in voice being changed.^[2]

Alongside regular text-to-speech functionality, support for pitch and rhythm transfer was added on June 30, 2021. Voices supporting this function can copy the intonation of a selected or uploaded reference audio clip.^{[t 1]}

Uberduck collaborated with visual artist JIMMY and rapper B la B to create a non-fungible token titled "UPLOADED", minted on September 20, 2021, consisting of a track voiced by a synthesized version of the rapper and accompanied by AI-generated visuals.^{[t 2] [7]}

The video generation aspect of Uberduck was promoted during Y Combinator's Winter 2022 Demo Day, where it was described by the founders as "Canva for programmable video".^[6] [i'd mention the yotta savings collaboration before this but i can't find confirmation from a non-uberduck party that uberduck was responsible for it]