EVERY SONG HAS A READOUT.
FIVE BUCKS. FOREVER.

Drop a track. Hear free songs that sound like it. Use them in your YouTube videos, podcasts, indie games, Twitch streams, and DJ sets: no copyright strikes, no subscription, $5 forever.

Under the hood: 13 audio traits (BPM, musical key, LUFS loudness, energy, mood, valence) and 20 closest matches from a curated catalogue of 3,382 Creative-Commons tracks. All commercial-safe. CC0, CC BY, or CC BY-SA only: never CC BY-NC. Source: Free Music Archive, ccMixter, Jamendo.

What everysong is for

• YouTube creators picking music for video essays, monetised channels, shorts, vlogs, tutorials. Matches won't trigger Content ID when used per the license terms.
• Podcasters finding intros, beds, transitions, and ad-spot music without paying ASCAP/BMI/PRO fees.
• Indie game developers scoring level themes, ambient beds, combat tracks, menu loops. Ships safely in commercial games on Steam, itch.io, GOG, Epic, consoles.
• Twitch streamers finding DMCA-safe background music for live broadcasts and VODs.
• Music producers A/B-ing work-in-progress against the trait fingerprint of a reference song.
• DJs finding CC-licensed crate-fillers that match the energy and key of existing sets.

How it works

1. Upload any audio file (MP3, WAV, FLAC, M4A, or OGG, up to 30 MB).
2. Wait 10-15 seconds. librosa extracts low-level signal features, CLAP generates a 512-dim audio embedding.
3. Read the 13-trait card (BPM, key, LUFS, energy, valence, danceability, acousticness, instrumentalness, plus 5 more: each labelled GREEN or AMBER for reliability).
4. Pick from 20 ranked CC matches: each with a license badge and a deep link to the source page on Free Music Archive, ccMixter, or Jamendo.
5. Use the match in your project. Credit per license terms. Ship.

Full pipeline walkthrough: how everysong works.

Pricing

$5 USD, paid once, for lifetime access. No subscription, no per-track fees, no upgrades. Fair-use cap of 50 uploads per day per account: message [email protected] if you need more. 30-day no-questions refunds at [email protected]. If the project shuts down, the analysis code goes open-source.

What it is not

• Not Shazam: it analyses audio characteristics, doesn't identify the song.
• Not a streaming service: returns links to original sources, doesn't host audio.
• Not a subscription: $5 once, forever.
• Not an AI music generator: returns real recordings by real humans.
• Not a label-tier audio AI like Cyanite or Musiio: this is the indie-tier $5 version.

Deep dives

• How it works: the 5-hop pipeline, the 13 audio traits, GREEN/AMBER reliability tiers
• For studios, labels, and platforms (B2B service tier)
• About everysong: the project, the promises, the infrastructure
• Free music for YouTube videos: without copyright strikes
• Free music for podcasts: intros, beds, transitions, ad spots
• Free music for indie games: level themes, combat, menus
• The blog: Creative Commons licensing, audio similarity, YouTube copyright
• 30-day refund policy
• Privacy policy
• Terms of service

Built and maintained by Zara, an independent developer in the Netherlands. Contact: [email protected]. If you're reading this, your browser doesn't have JavaScript enabled: the full interactive version of this page loads with JS. All product functionality (upload, analysis, payment) is available regardless.