search

suno-ai / bark

🔊 Text-Prompted Generative Audio Model
MIT License
35.31k stars 4.15k forks source link

13 second limit? #60

Closed 7gxycn08 closed 1 year ago

7gxycn08 commented 1 year ago

How do you unlock the limit?

wsippel commented 1 year ago

There's nothing to unlock - it's a limitation of the model itself, not an arbitrary limit set by Suno. To generate longer audio, you can split the input, generate audio chunk by chunk, then join the individual chunks. Comes with caveats but works pretty well with some speakers. That's how I've implemented it in the Oobabooga extension.

gkucsko commented 1 year ago

yup! people are writing great little tools to stitch things together. also see https://github.com/suno-ai/bark/issues/21