YouTube to MP3 (with Thumbnail & Metadata)

Extract audio from any YouTube video as MP3, with the video thumbnail embedded as cover art and ID3v2 metadata (title, artist, album, year, comment). Powered by yt-dlp + ffmpeg server-side. Single videos only — playlists and live streams are rejected.

Use responsibly: only download content you have rights to (your own uploads, Creative Commons licensed, public domain, or otherwise legal in your jurisdiction). Respect creators and YouTube's Terms of Service.

Use a browser extension like "Get cookies.txt LOCALLY" (Chrome) or "cookies.txt" (Firefox) on youtube.com while logged in, then paste here. Cookies are sent to this server only for the current request, written to a per-request temp file (chmod 600), and deleted as soon as yt-dlp finishes. They are never logged or stored.

Security tip: use a separate (throwaway) YouTube account, not your main one. Anyone with these cookies can act as that account until you log out of the browser you exported from.

Not provided — bot-check may apply

Frequently Asked Questions

What does "embed thumbnail" and "embed metadata" mean?

The MP3 you download gets a cover art image (the YouTube thumbnail) baked into its ID3v2 tags as an APIC frame — your music player will display it as the album art. The metadata tags include the video title, channel name (as artist), upload year, and a comment field with the original URL. Compatible with iTunes, Foobar2000, VLC, Plex, Jellyfin, and any modern player.

Which bitrate should I pick?

320 kbps for archival/audiophile use (≈ 2.4 MB per minute). 192 kbps is a good balance — near-CD quality at half the size, recommended for most users. 128 kbps is smallest, fine for speech/podcasts but audible artifacts on music. Note: YouTube's source audio is usually 128 kbps Opus or AAC — encoding above the source bitrate adds no quality, just bytes.

Is this legal?

Depends on the content and your jurisdiction. Legal use cases include: downloading your own uploads, Creative Commons / public-domain videos, content where the creator has explicitly permitted downloads, and personal-use copies in jurisdictions that allow private copying. Mass downloading of copyrighted content for redistribution is illegal in most places and violates YouTube's Terms of Service. The tool is provided for legitimate use — you're responsible for the URLs you submit.

Why are playlists / live streams rejected?

Playlists can hold thousands of videos — one click would tie up server resources for hours. Live streams have no fixed end, can't be "downloaded" as a complete file, and would consume bandwidth indefinitely. Both are explicitly rejected. The URL validator only accepts single-video URLs (youtube.com/watch?v=..., youtu.be/..., or shorts/embed equivalents).

What's the file size limit and rate limit?

Output is capped at 150 MB per download (covers everything except hours-long videos at 320 kbps). Per-IP rate limit is 5 downloads per 10 minutes — refresh-spammers hit a 429 response. Server-side timeout: 5 minutes per request.

I get "Sign in to confirm you're not a bot" — what now?

YouTube increasingly blocks server / datacenter IPs unless cookies from a logged-in browser session are provided. The tool surfaces this error and auto-opens the Advanced: paste cookies panel.

How to get cookies (Chrome / Brave / Edge):

  • Install the "Get cookies.txt LOCALLY" extension
  • Open youtube.com and make sure you're logged in
  • Click the extension icon → "Current Site" → "Export"
  • Copy the contents of the downloaded file and paste into the textarea here

Firefox: install "cookies.txt" from add-ons, same workflow.

Two accepted formats: the full Netscape cookies.txt file, OR a header-style key=value; key=value; blob (the tool converts it). Cookies are written to a per-request temp file (chmod 600), passed to yt-dlp via --cookies, and deleted with the rest of the work dir as soon as yt-dlp finishes. They're never logged, never stored, never reused for another request.

Important: use a separate / throwaway YouTube account, not your main one. Anyone who intercepts these cookies could act as that account until you log out.

Is my IP / URL logged?

Only minimally: the rate-limiter tracks request timestamps per IP in memory (in-process Map, cleared on restart, never written to disk). The URL itself isn't logged beyond that. Downloaded files are written to a temp directory, streamed to you, and deleted as soon as the response closes.