illscience / dreaminaudio

Fine-tune of music gen with tracks from my record label Dream In Audio.

  • Public
  • 127 runs
Run with an API
Playground API Examples README Versions

Run time and cost

This model costs approximately $0.23 to run on Replicate, or 4 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia A100 (80GB) GPU hardware. Predictions typically complete within 3 minutes. The predict time for this model varies significantly based on the inputs.

Readme

This is a fune-tune of music gen with music from my 2011 record label Dream In Audio. You can hear some of the tracks from the label here: https://soundcloud.com/dreaminaudio

The sounds is “west coast house” - a genre that had its hey day in the 1990s San Francisco sounds and parties pioneered by DJ Mark Farina.

The model is still WIP - would love any advice that folks have on how to best label the training data (i.e. using essentia vs hand labeling) and also how to evoke elements from specific tracks in the prompts.