Selectia is a music player and library manager for DJs. It feature exeperimental features to help manage and play tracks in advanced ways.

Commonon features such a tempo and beat grid analysis are still WIP but with promising results. With the current technis it also possible to detect events within a track such as beats, bars, drops, etc... As the audio analysis is done by spliting the audio into stems (i.e vocals, drums, bass, etc...) we should be able to implement some interesting features.

• Generate midi notes based on audio stems (i.e generate a midi file for drums, bass, vocals etc)
• Apply FX based on audio stems (i.e apply a reverb to vocals only)

Selectia can load directories, scan audio files, index them and store the metadata in a database. It handle doublon seamlessly and handle removal of files that are no longer in the library as well.

The tagging feature is at the core of Selectia. It allows you to tag your music files with metadata such as artist, album, genre, bpm and more. The tagging system is also used for search, filtering and recommendation.

When you add a new music file to your library, Selectia can automatically tag it using a combination of pattern matching and AI. Selectia can also suggest tags based on the file name, the file path, the audio data and existing tags to allow fast and non repetitive tagging.

• Node.js (>22 recommended)
• Yarn
• Rust nightly
• SQLite system library
• Tensorflow >2.x Note: Tensorflow library must be reachable from node install directory to work in dev mode prebuilt tensorflow.dll can be found in packages/selectia-app/src-tauri/resources/tensorflow.dll

cd packages/selectia-app
yarn install

cd packages/selectia-app
yarn tauri dev

The Rust/Typescript bindings are generated using ts-rs. You can regenerate them by running cargo test -p selectia-app which will generate the src-tauri/bindings folder.