Audio tools

Every tool here follows the same rule, and it is the rule most audio websites break: nothing you feed in is ever compressed a second time. One of these tools takes a soundtrack out of a video without decoding it at all. The other five decode the audio, do the work on the samples, and write the result out uncompressed. Getting this wrong is the single most common way people quietly damage their own recordings, so it is worth a page explaining it properly.

Lossy, lossless, and why a second pass is the mistake

Formats split into two families. Lossless ones — WAV, FLAC, ALAC — store the waveform so that decoding gives back exactly what went in. Lossy ones — MP3, AAC, Opus, Vorbis — throw away detail a model says you will not notice, and the discarded part is gone permanently.

The trouble starts when a lossy file is decoded and compressed again. The second encoder sees the first encoder’s artefacts as real signal and spends bits preserving them, while making its own fresh set of compromises on top. The result is measurably worse and often audibly worse, especially in cymbals, sibilance and reverb tails. It also does not save space in any useful way: to hide the damage you have to pick a bitrate high enough that the file is no smaller than the original.

So the rule is simple. If audio is already lossy and you are not editing it, do not re-encode it. Move it, copy it, rewrap it — but do not run it through an encoder to change the file extension. Converting to a lossless format afterwards does not help either: it faithfully preserves the damage in a larger file.

Why extracting a soundtrack should never re-encode it

Nearly every video you will encounter carries AAC audio. That track was encoded once, when the video was recorded or exported, and that encode is the best version of the sound that exists anywhere. Copying it into its own container gives you exactly that, in milliseconds, with no settings to get wrong.

Most sites that offer this immediately produce an MP3 instead, because MP3 is what people search for. That is a second lossy pass applied for the sake of a filename. The output is an approximation of an approximation, and the original that could have been kept intact is gone unless you still have the video.

Our extract tool writes an M4A file: the same MP4 container you already had, holding only the audio track, byte for byte. It reports how many audio packets it copied so you can confirm nothing was touched. Nothing is uploaded — which matters more here than in most categories, because extracted audio is often the most sensitive part of a recording.

Bitrate, sample rate and container: three things people mix up

Bitrate is how many bits per second the compressed stream uses. It is the main quality dial on a lossy encode, and it is a property of an encode that has already happened — you cannot raise it after the fact. Re-encoding a 128 kbps file at 320 kbps produces a bigger file containing the same damaged audio, which is one of the most common wasted operations in audio.

Sample rate is how many times per second the waveform was measured — 44.1 kHz for CD, 48 kHz for most video. It sets the highest frequency that can exist in the file at all, and like bitrate, raising it afterwards adds nothing. Bit depth (16-bit, 24-bit) is a third, separate thing and only applies to uncompressed audio; a lossy file does not have one in any meaningful sense.

The container is the wrapper, and it is not the codec. An .m4a file holds AAC; so does an .aac file and, usually, the audio inside an .mp4. A .wav can hold uncompressed PCM or, awkwardly, compressed data. This is why converting to M4A and converting to AAC can mean the same thing or different things depending on what you started with, and why a tool should tell you the codec rather than only the extension.

What is here, what it costs, and what we will not pretend

Two different kinds of tool live here, and the difference matters. Extracting a soundtrack from a video never decodes anything — the compressed frames are copied across untouched, so the result is bit-for-bit the audio that was in your video. Trimming, joining, changing volume and converting to WAV all have to decode, because you cannot cut or scale compressed frames without turning them back into samples first.

For that second group we had a choice on the way out: compress again, or store the samples as they are. We store them as they are — 16-bit PCM in a WAV — so no second lossy pass ever happens. That turned out not to need an encoder at all, which is why these tools have no external dependencies and no WebAssembly: your browser already contains the decoder, and a WAV header is 44 bytes we write ourselves.

The cost is size, and it is not small: uncompressed audio at 44.1 kHz in stereo is about 10.1 MB per minute, so a four-minute clip from a 4 MB MP3 comes out around 40 MB. Every tool shows the exact figure before you press anything. That is the honest trade, stated rather than buried.

And here is the line we will not cross, including when it applies to us. We do not call this lossless. No second lossy compression happens, which is a real and checkable guarantee — but if your source was an MP3 or an AAC track, the loss happened when it was first encoded and decoding cannot undo it. Calling an uncompressed output lossless when the input was lossy is exactly the kind of half-true label this page exists to argue against.

There is still no MP3 export, and that is the same decision rather than a gap. Encoding MP3 from audio that is already lossy is a second lossy pass, and it needs a page that says so plainly, along with an encoder whose licence we are willing to ship. It is planned as its own tool, not as a hidden option inside one of these.