How to Extract Audio from a Video (And Why 320 kbps Is Usually a Mistake)

Getting the sound out of a video, what MP3, M4A and AAC actually are, and why the highest quality setting nearly always makes a bigger file that cannot sound better.

Extracting audio from a video means taking the sound track out and saving it as its own file. Every video is really two streams travelling together — pictures and sound — and pulling one out is a matter of copying it, not of converting anything.

That distinction is the whole subject, and it is where most tools go wrong.

How do I extract audio from a video?

Open the video in a tool that can read it, choose whether you want the audio copied out as it is or re-encoded as MP3, and save the result. It takes seconds.

The four steps, whatever you use:

  1. Add the video. MP4, MOV, MKV, WebM and AVI all carry audio the same way.
  2. Decide between copying and converting. Copying is faster and loses nothing. Converting to MP3 gives you a file that plays on absolutely anything.
  3. Pick a quality, if you are converting. More on this below, because it is the step everybody gets wrong.
  4. Save it.

Lonecto's video to MP3 tool does this in your browser — the video is never uploaded, there is no account, and it reads the file first so it can tell you what the audio in it actually is.

What is the difference between copying and converting the audio?

Copying takes the existing audio track out of the container untouched. Converting decodes it and encodes it again as something else, which always loses a little quality.

The difference in practice is larger than it sounds:

Copying is right whenever you do not specifically need an MP3. Converting is right when you do.

What bitrate should I use for the MP3?

One step above whatever the video already contains — which for most video means 192 kbps, not 320.

Here is the part almost nobody explains. The audio inside a video has already been compressed once. A typical MP4 carries around 128 kbps of AAC: good enough that you will never notice, and permanently missing whatever the first encoder discarded to get there.

Converting that to a 320 kbps MP3 does not bring any of it back. You get a file roughly two and a half times the size of the audio it came from, which at best sounds identical. The extra bits are spent describing the gaps the first encoder already made.

So:

Encoding to exactly the source bitrate is the other common mistake: the second encoder makes different decisions from the first, and at the same budget those differences become audible. One step up is the compromise.

What is bitrate?

Bitrate is how much data is spent on each second of audio, measured in kilobits per second. Higher means a bigger file and, up to a point, better sound.

The "up to a point" matters. Bitrate is a budget, not a quality score. Spending more than the source material contains cannot add detail — it only records the existing detail more precisely, including the parts that were already thrown away.

For reference: 128 kbps is the audio in most video, 192 to 256 is where most music streaming sits, 320 is the top of the MP3 scale, and uncompressed CD audio is 1,411 kbps.

MP3 vs M4A: which should I use?

M4A is the better format and MP3 is the more compatible one, so the answer depends entirely on where the file is going.

Choose M4A when the file is for you, your phone, or anything modern. Choose MP3 when you are handing it to someone else and cannot be sure what they will open it with, or when a piece of software specifically asks for one.

AAC vs MP3: what is the actual difference?

AAC is the successor to MP3 and is meaningfully more efficient — roughly, AAC at 128 kbps is comparable to MP3 at around 192 kbps.

The confusion here is usually about names. AAC is a codec: the method of compression. M4A is a container: the file that holds it. So "AAC vs MP3" is a comparison of codecs, and "M4A vs MP3" is very nearly the same comparison wearing different clothes, because MP3 is unusual in being both a codec and a file format at once.

AAC is what YouTube, Apple Music, Spotify and virtually all video use. MP3 survives on compatibility rather than merit, and there is nothing wrong with that — a format that plays everywhere has real value.

Does extracting audio reduce the quality?

Only if you re-encode. Copying the existing track out is lossless, because nothing is decoded or re-compressed — the same data is simply written into a different file.

If you do convert, you are compressing something that was already compressed, which is called generation loss. It is usually inaudible at a sensible bitrate and it is never zero. The way to avoid it entirely is to copy instead.

Can I extract audio from a video on my phone?

Yes. A browser-based tool works on a phone as well as a laptop, because the work happens on the device either way — the only difference is that a phone's processor is smaller, so a very long video takes longer.

There is no need to install anything, and no need to send a personal video to a stranger's server to get the sound out of it.

How do I get just part of the audio?

Set a start and an end before converting, so only that section is processed. One song out of a long recording, one answer out of an interview, one section of a lecture.

This is also considerably faster than converting the whole file and trimming afterwards, because a good tool will skip straight to the part you asked for rather than decoding everything before it.

Is it legal to extract audio from a video?

It depends entirely on the video, and the honest answer is that this is a question about copyright rather than about software.

Extracting the audio from something you recorded, something you were sent, or something you have permission to use is straightforward. Taking the audio from someone else's copyrighted work and redistributing it is not, whatever tool was used. Rules vary by country and by what you intend to do with it, and a blog post is not the place to be confident about your particular case.

What is worth knowing is that a tool which only accepts a file from your device is a different proposition from one that fetches from a streaming platform. The first works on material you already have. The second is doing something else.

Questions

What format should I extract audio to? M4A by copying, if you have the choice — it is faster and loses nothing. MP3 if the file is going to someone else or into older equipment.

Why is my extracted MP3 bigger than I expected? Because you chose a bitrate above what the video contained. A 320 kbps MP3 of a 128 kbps source is 2.5 times the size and no better. Drop to 192.

Can I extract audio without losing quality? Yes — copy the existing track rather than converting it. Nothing is re-compressed, so nothing is lost.

How long does it take? Copying is close to instant regardless of length; measured at about a second for a 55-minute video. Converting to MP3 took about 34 seconds for the same file on a desktop, and longer on a phone.

Does the video file get changed? No. The original is read and left exactly as it was; the audio is written to a new file alongside it.

Can I extract audio from a YouTube video? Not with a tool that only takes files from your device, which is deliberate. Downloading from streaming platforms is a different activity with different rules attached, and it is worth being clear about which one you are doing.

What about video with no sound? Some videos genuinely have no audio track — screen recordings often do not. A good tool will tell you that before you wait, rather than producing an empty file.