When you’ve ever dreamed of creating songs with Tupac or Jay-Z, an AI software known as Uberduck can take you near fulfilling your fantasies.
Uberduck is considered one of a variety of instruments that permits you to select a star voice after which enter textual content for them to talk. What units it other than the others is it could actually do a reasonably spectacular job of replicating a rapper’s stream.
You’ll be able to synthesize speech right into a “calm” or “intense” Tupac verse, for instance, or attempt an Eminem “freestyle” verse or “pre-Eminem Present” stream.
Uberduck’s creator says they began engaged on the system with the objective of making an open-ended dialog agent:
I constructed an interactive audio chatbot over WebRTC that generated textual content responses with a Transformer mannequin and synthesized them to audio, however I discovered that speech synthesis was essentially the most thrilling a part of the mission.
The software blew up on TikTok after a lawsuit pressured the app to swap its text-to-speech voice for a special model. Many customers have been unimpressed by the substitute and made the swap to Uberduck.
[Read: Why entrepreneurship in emerging markets matters]
They’ve gone on to make use of the software in a variety of inventive methods, from including Biggie verses to their very own tracks to creating Kanye West rap the lyrics to Bohemian Rhapsody.
Even Linkin Park’s Mike Shinoda has tried it out:
@mshinodaHappy Endings by me feat. Infamous B.I.G. ##Biggie##massive##HappyEndings##Ai##Uberduck##texttospeech##deepfake##producer##music##rappersoftiktok##fyp♬ unique sound – Mike Shinoda
When you’d quite have a dialog together with your favourite MC — or drive them to provide you a shout-out — that’s additionally doable.
The synthesized voices are removed from good, however with tweaks to the textual content — including additional vowels, for example, to increase a syllable — you may generate some fairly correct imitations of your favourite rappers. I’m fairly happy with these Tupac bars I created:
To my ears, the voice is more convincing than the Faux-Pac from this classic Dave Chappelle skit:
The tool does, however, have some potential to be used for disinformation and defamation.
Uberduck’s terms attempt to allay these concerns. Users are prohibited from using the outputs for commercial purposes or the production of defamatory material. They must also clearly identify that their creations were generated by AI, and will be banned if they violate the rules. In addition, the tool’s inventor says they’ll remove voices from the site upon request.
Putting words into a replica of someone’s voice without their permission could also be viewed as disrespectful, but these feel more like impersonations than recreations.
HT — mrcomposition09
Greetings Humanoids! Did you know we have a newsletter all about AI? You can subscribe to it right here.