How to Convert YouTube Videos to MP3: The Complete Guide
Pulling the audio out of a video is one of the most common things people want from a downloader, and one of the least well explained. Most guides tell you to paste a link and click a button, which works, and then leave you with no idea why the file sounds the way it does or how large it should have been.
This covers the whole picture: what the tool is actually doing, what quality you can reasonably expect, how to size the files, and where the legal line genuinely sits rather than where people assume it does.
Extraction is not conversion
These get used interchangeably and they are different operations with different costs.
A video file is a container holding a video stream and an audio stream side by side. Extraction pulls the audio stream out and writes it to its own file without touching the data. It is a copy operation. It takes about a second and loses nothing.
Conversion re-encodes that audio into a different codec. YouTube stores AAC or Opus, and MP3 is neither, so producing a genuine .mp3 file means decoding and re-encoding. Every re-encode of lossy audio loses a little more, in the same way photocopying a photocopy does.
This matters for one practical reason: if your destination accepts M4A, which holds AAC, take that instead of MP3. It is the untouched original stream and it is smaller at the same quality. MP3 is worth choosing only when something in your chain insists on it, which these days is mostly older car stereos and a few DJ applications.
What bitrate you actually get
Here is the number that sets the ceiling: YouTube serves audio at roughly 128 kbps AAC for most videos, and around 160 kbps Opus for the WebM streams. Music-heavy content on some channels goes a little higher. That is the source.
- Extracting to M4A preserves whatever YouTube stored, typically 128 kbps AAC.
- Converting to 128 kbps MP3 gives you a file that sounds slightly worse than the source, because it was re-encoded.
- Converting to 320 kbps MP3 gives you a much larger file that still sounds like the 128 kbps source. The extra bits describe the artefacts more precisely rather than restoring detail.
- Nothing anywhere produces a file better than what YouTube holds.
File sizes you can plan around
Audio bitrate translates directly into file size, which makes this easy to estimate. At 128 kbps you are looking at roughly 1 MB per minute.
- A 3 minute song: about 3 MB.
- A 10 minute video essay: about 9 MB.
- A 45 minute interview: about 40 MB.
- A 2 hour podcast: about 115 MB.
- The same 2 hour podcast as 720p video: somewhere between 700 MB and 1.5 GB.
That last comparison is the real argument for audio extraction. For anything you are only going to listen to, keeping the video is ten times the storage for a picture you will never look at.
Doing it
- Copy the video link. A youtube.com/watch?v= URL, a youtu.be short link and a /shorts/ link all work and all point at the same media.
- Paste it into the YouTube to MP3 converter.
- Wait for it to read the video and list what is available.
- Pick the audio option. Save.
On desktop the file lands in Downloads. On Android it goes to the Downloads folder and most music apps will index it if you move it into your music directory. On iPhone it goes into Files, and note that Photos will not accept an audio file, so it stays there. That is correct behaviour rather than a failure.
Tagging, which nobody mentions and everybody needs
An extracted file arrives with essentially no metadata. No artist, no album, no track number, often a filename that is a video ID. Drop a folder of those into a music app and you get a wall of untitled tracks.
Fix it at the point of download rather than later. Rename the file to something with the artist and title in it while you still remember what it is, and if you are building a library, run the files through a tagger such as MusicBrainz Picard or the tag editor in your music app. Ten seconds per file now saves an afternoon in six months.
When a download comes back silent
Occasionally an audio extraction produces nothing, or a video download you expected to have sound turns out to be silent. There are four causes and only one of them is a fault.
- Above 1080p, YouTube stores video and audio as separate streams. Taking the video stream alone gives a silent file. Nothing is broken.
- The uploader muted the original, sometimes to dodge a copyright claim on background music.
- The platform stripped the audio retroactively after a rights claim. A video that had sound last year can be silent now.
- Your player cannot decode Opus, which is what WebM files carry. VLC will play it; some older software will not.
Our piece on why downloaded videos have no sound walks through diagnosing each one in order.
The legal part, stated plainly
This gets hand-waved constantly, so here is the honest version.
YouTube's terms of service prohibit downloading content except through features YouTube provides, such as its own offline mode in the app. That is a contract between you and YouTube. Breaching it is not a criminal matter, but it is a breach.
Copyright is the separate and more serious question. The music in a video is owned by whoever wrote and recorded it, and that ownership is unaffected by the video being publicly viewable. Many countries allow a private copy for personal use and many do not, and the rules vary enough that a blanket statement would be wrong.
What is clearly on the wrong side of the line anywhere: re-uploading extracted audio, using it in monetised content, distributing it, or putting it in a commercial project. What is generally treated as ordinary personal use: keeping a copy of something you already have the right to listen to, extracting audio from your own uploads, or saving a lecture, a public domain recording or a Creative Commons track.
If you want music you can keep, a streaming subscription with offline playback costs less than most people spend on coffee in a week and removes the question entirely. Our disclaimer sets out where we stand on all of this.
The same job on the other platforms
Audio extraction works identically everywhere, with different source bitrates.
- Twitter to MP3 pulls the audio from X posts. Note that X GIF style clips have no audio track at all, so there is nothing to extract.
- TikTok to MP3 handles TikTok sounds, which are usually short and heavily compressed.
- Instagram and Facebook audio can be extracted from the downloaded MP4 locally with ffmpeg if you need it.
For the video side, the YouTube downloader lists every rendition a video holds, and MP4 vs WebM explains the container choice you will be offered on the higher resolutions.
YouTube serves audio at roughly 128 kbps AAC, or around 160 kbps Opus for WebM streams. That is the ceiling. Converting to a higher MP3 bitrate produces a larger file, not a better one.
No, when the source is 128 kbps. The extra bits describe the existing compression artefacts more precisely rather than restoring lost detail. You get two and a half times the file size for no audible improvement.
M4A if your destination accepts it. It holds the untouched AAC stream YouTube stored, so nothing is re-encoded, and it is smaller at the same quality. Choose MP3 only when something in your chain requires it.
Roughly 1 MB per minute at 128 kbps. A three minute song is about 3 MB, a 45 minute interview about 40 MB, and a two hour podcast about 115 MB. The same podcast as video would be ten times that.
YouTube's terms prohibit downloading outside its own offline feature, and separately the music is owned by whoever made it. Personal copies are treated differently in different countries. Re-uploading, distributing or monetising extracted audio is clearly infringing everywhere.
Either the video itself was silent, the platform stripped the audio after a rights claim, or your player cannot decode Opus. Open the original video with the volume up first: if it is silent there, no tool can add audio that was never recorded.