Generative AI startup Uncharted Labs has launched Udio, a text-to-music generator that generates high-quality music from text prompts.
A number of weeks ago, Gen AI enthusiasts were raving in regards to the music Suno V3 was producing. User feedback from Udio's early testing suggests that Suno V3's time at the highest can have been short-lived.
Udio is like ChatGPT for music. You bring it up with an outline of what the song is speculated to be about, in addition to a kind of music, and click on the perhaps paradoxically labeled “Create” button.
Udio mechanically generates lyrics to match your latest creation, but in addition means that you can write your personal lyrics in English, Chinese, Japanese, Russian, Polish, German, French or Italian.
The tool generates a 30-second clip in just a few seconds, which you’ll be able to extend to 90 seconds and edit on the platform.
Tracks generated using early AI music tools were fairly generic, repetitive, and fairly easy. Udio's musical and vocal output is complex and sounds nearly as good as much business music produced by human artists.
It won't make you hand over your Pink Floyd or Led Zepp albums, however the indisputable fact that music like this is totally AI generated is crazy.
Crater lullaby pic.twitter.com/2jZGv7TfIw
– David M. Comfort (@DavidmComfort) April 10, 2024
Training data and superb print
Uncharted Labs' founders include former Google DeepMind engineers. The company raised $10 million in seed funding from investors including Andreessen Horowitz's a16z and Instagram co-founder Mike Kreiger.
William from Blackeyed Peas is considered one of the notable musicians who’ve invested in Udio. He has been outspoken in support of AI music, a view not shared by the 200 musicians who signed an open letter last week denouncing AI's involvement in music production.
Udio's developers haven't revealed the small print of their training data, however the tool was almost definitely trained on copyrighted music.
Udio's website states: “Udio doesn’t generate songs using artists' voices.” That could also be so, however it appears like the model spent a number of time listening to human artists.
Rolling Stone's Brian Hiatt noted the familiar sounds of Tom Petty's voice in a number of the songs generated.
Udio, the brand new Suno competitor, generated songs for me twice with Tom Petty's voice. https://t.co/DJxnVjQaNT
— Brian Hiatt (@hiattb) April 10, 2024
The superb print in Udio's terms and conditions encourages anyone dissatisfied with its service to first initiate an “informal procedure” quite than taking legal motion. I believe there’ll soon be more copyright infringement class actions much like those involving OpenAI, Meta, and Microsoft.
share is free through the beta testing period and allows users to create as much as 1,200 songs per 30 days that they’ll use for business purposes.
If you're making a catchy tune with Udio, do not forget that you continue to don't have copyright on AI content within the US. And perhaps you are feeling a bit of confused whenever you tell someone, “Hey, take heed to this song I made.”
