← Latest products

A lightweight ML model to predict music emotion

energy, valence, etc. A Python tool and Beets plugin for music emotion and perceptual feature prediction (valence, energy, danceability, etc.). Spotify has 7 features for each of their music tracks (acousticness, danceability, energy, instrumentalness, liveness, speechiness, valence) which describe the perceptual/emotional content of the song.

Open source

77 stars · GitHubexternal site

Visit github.comexternal site

Reactions (1)

Similar products

by description similarity

  1. Mixlist

    mixlist.org

    built a web app that uses k-means clustering on artist genres (one or multiple) to automatically organize Spotify liked songs into playlists.

    Hacker News33 pts

  2. Spotify Blend + Mood

    ripple.orcas.land

    Hacker News6 pts

  3. AI audio embeddings

    speakmusic.sonophase.com

    Discover 100M+ iTunes songs with language models

    Hacker News7 pts

  4. AI-Powered Music Creation Starts Here

    vibemusicing.com

    Vibe Musicing

    Hacker News6 pts

  5. Free AI-based music demixing in the browser

    sevag.xyz

    sevagh's blog

    Hacker News190 pts