Analyzes music to determine song structure, bpm, downbeats, and demuxes audio
This model costs approximately $0.098 to run on Replicate, or 10 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 A40 (Large) GPU hardware. Predictions typically complete within 136 seconds. The predict time for this model varies significantly based on the inputs.
Cog implementation of all-in-one forked from: https://replicate.com/sakemin/all-in-one-music-structure-analyzer
All-in-one (Taejun Kim): https://github.com/mir-aidj/all-in-one
This implementation differs by optionally exporting demuxed audio files.
