Martin Camacho
Martin Camacho
Hi all - great library! I just wanted to second the CSS-in-JS comments above and share a couple CSS-in-JS approaches to subcomponent styling in the wild: 1) [react-select](https://react-select.com/styles#style-object) 2) [Chakra...
Sorry for the delay. Unfortunately, not currently - it's definitely a more complex problem - but happy to think through approaches.
Working on in #8 - I need to refactor some of the click logic. Clicking on a mark currently deselects it
The best way right now is probably to split text up into sentences or chunks, and generate separately, passing in the same speaker prompt for consistency. We could consider a...
@AigizK just so we know - which language are you interested in? Thanks!
Do you have a prompt/sample you can share? 24khz could definitely be one reason
You can try ``` [door shutting] ``` as a prompt. Sometimes works better if you add more text ``` [door shutting] Hello world, that was a sound of a door...
Hi, those progress bars are the models being downloaded (this happens once, and they are 5-10GB total). So audio generation is not happening yet.
Re Apple Silicon, definitely something we welcome input/suggestions on. Re where results end up, they start in memory. For example saving to wav/mp3 check out https://github.com/suno-ai/bark/issues/13
In our experiments with this set of models, for more interesting tokens such as `♪`, you might need several generations to get something very good, but the good generations are...