Guide · 2 min read
What Daydream does
Daydream is the first AI-native instrument. Leveraging cutting-edge audio synthesis research, it delivers near-real-time controllability of audio generation. While it's registered as an Effect in most DAWs, it behaves more like an instrument: using any input, it continuously generates music based on your input.
Daydream takes input audio — a loop, a stem, a sample, a full mix, or even white noise — and generates brand-new audio. Depending on your settings, you may generate audio that incorporates elements of your source audio, or fabricate from scratch using prompts and parameter updates. Generation typically runs on Daydream's servers and streams back into your session, and a local option is available for those who have access to a GPU.
Things to know
- Daydream is registered as an effect in most DAWs
- While Daydream behaves like an instrument in many ways, we've registered it as an effect because it does require input to function. If this poses problems for your workflow, please let us know.
- Output is audio, not MIDI
- If you need editable notes, run the output through a separate pitch-to-MIDI step.
- Output is a full mix
- Solo instruments coming soon.
- The output is yours
- Creating music with Daydream involves human skill beyond prompting; the audio you generate with Daydream is your creative output and belongs to you.