Traditionally, music producers sourced samples by visiting record stores, flea markets, and thrift shops to seek vinyl records that might contain sampling material. This meticulous process, known as crate digging, involves physically searching through crates of vinyl records to find musical sections that can be reused in new compositions.
With the proliferation of digital music access, sample libraries, and online resources, digital crate digging has emerged as the contemporary alternative. Producers can now explore platforms like Splice and Tracklib to find suitable sampling material for their productions. However, the digitisation of music has introduced new challenges in navigating and managing these extensive collections of musical material.
Over time, producers accumulate vast libraries of samples and music, often with inconsistent or non-existent labelling schemes. This can make it difficult to search for specific samples using traditional search methods. Finding the right sample can be a daunting task, often requiring hours of manually listening through tracks.
Crate-Dig has been designed to streamline the digital crate digging experience, allowing producers to search their music and sample collections using natural language or audio input. By leveraging advanced AI search capabilities, Crate-Dig helps discover hidden gems that might have been forgotten or overlooked, saving countless hours and enhancing the creative workflow.
Download the latest release from the releases page, unzip the folder and run the executable.
- Pick a folder with your audio library in it, this could be something like a folder you keep all your sample packs in or a folder where you download all your tracks and albums to.
- Then click analyze and the tool will extract information about your tracks and store them in a db (UserLibrary/embeddings/embeddings.npy)
- Once the analysis is complete you can search for a track by typing in the search bar and clicking search. The tool will then find the most similar tracks to the one you searched for and display them in the UI and save a playlist file in the UserLibrary/playlists folder.
Updates tagged as "drop-in" will be able to be extracted over the top of the existing installation and would be very lightweight (~40MB), these will be the executables only and would not include the _internal folder. Each release will have a drop-in update available.
Updates tagged as "full" will require the user to delete the existing installation and extract the new version.
Pull the repository and run pyinstaller cratedigAI.spec in the root directory of the project. This will create a dist folder with the executable and _internal folder that has the modules needed to run the executable.
Alternatively you can run pyinstaller cratedigAI_exec.spec to create a one file executable vs the one folder executable, one file executables are larger, easier to distribute but take longer to start up.