Sonoteller is an AI-powered analysis tool that listens to a song and generates a structured breakdown of both its lyrics and its music. Instead of manually tagging tracks for a catalog, playlist, or distribution platform, users submit a song and receive an automated summary covering genre, subgenre, mood, instrumentation, BPM, key, and vocal characteristics on the music side, alongside a lyrical summary, themes, mood, language, and an explicit-content flag on the lyrics side. The tool also flags the golden minute of a track, the section containing the chorus or other highlight moment, which is useful for building previews or promotional clips without listening to a full song end to end.
The tool is built for people who work with large volumes of music and need consistent, searchable metadata rather than subjective tags. That includes music catalog managers, distributors, playlist curators, sync licensing teams, and developers building music discovery or recommendation features. Analysis typically takes about a minute to process both the lyrical and musical dimensions of a track, which makes it practical for batch-style workflows once integrated through the API rather than the one-song-at-a-time web demo.
In practice, the public web version lets visitors try the analysis on a curated set of YouTube video examples, since the underlying API is designed to work with music files the user actually owns. Teams that want to run their own catalog through Sonoteller need to reach out for API access, and the site explicitly invites inquiries with an offer of a discount for qualifying use cases, suggesting API pricing is negotiated rather than published as a fixed rate.
Sonoteller sits in the music autotagging and audio-intelligence space alongside tools used for content moderation, mood-based playlist generation, and metadata enrichment for streaming and licensing platforms. Its differentiator is combining lyric-level semantic analysis with music-level acoustic tagging in a single pass, rather than treating lyrics and audio as separate tools. The product is explicitly labeled as beta, so buyers evaluating it for production catalog work should expect some inconsistency while the underlying models mature.