Extract audio from video means separating a video file's soundtrack from its picture and saving it as a standalone audio file. The source container holds a video track and one or more audio tracks multiplexed together; extraction unpacks the audio track and writes it to a new file so the sound can be played, edited, transcribed, or reused without the original visuals. The result is typically a compressed audio container such as Opus inside WebM, MP3, WAV, or AAC, depending on the tool or codec used. Extraction is not the same as converting a video to a different video format, nor is it the same as recording audio from a speaker or screen — it copies an existing audio track out of a video container. The phrase covers any method that pulls that track out: desktop editors, command-line tools like FFmpeg, desktop apps, mobile apps, or a browser-based tool that decodes the video locally. Understanding this distinction matters because the output format, file size, and quality all depend on what tool you use and whether the work happens on your device or on a remote server.
What the Phrase Actually Describes
The phrase "extract audio from video" describes a single conceptual operation: pulling the audio stream out of a video file so it can live on its own. A video file is a container — usually MP4, MOV, MKV, WebM, or another format — that bundles separate tracks together. One track carries the moving images, another carries the sound, and a third may carry subtitles or chapter markers. The container's job is to keep those tracks synchronized on a shared timeline; the audio track itself is encoded with a codec such as AAC, MP3, Opus, or Vorbis.
When a tool extracts audio, it decodes the original audio track and writes it into a new audio-only file. The video track is discarded in that new file. The output file may use the same codec as the original audio, or it may be re-encoded into a different codec. That choice is what determines whether the output matches the source's audio quality or is a fresh compression.
This is why the term shows up alongside related labels such as "rip audio," "demux," or "audio extraction." Each word refers to the same idea with a slightly different flavor: ripping implies copying from a disc, demux is the technical term for splitting tracks out of a container, and extraction is the broadest label used by everyday software. Each path ends at the same place: one audio file that used to live inside a video.
Why Extraction Differs from Recording or Conversion
Three operations get confused with extraction, but each one produces a different result.
- Extraction copies an existing audio track from a video container into an audio-only file. The source audio is already there; the tool simply pulls it out.
- Recording captures sound through a microphone or system audio device as it plays. The source audio is created during the process, not pulled from a file.
- Conversion rewrites a file from one encoding to another. A video-to-video conversion keeps the picture; a video-to-audio conversion is a specific kind of conversion that includes the extraction step.
Extraction is the only one of the three that starts with an audio track already embedded in the video. Recording works on whatever the microphone hears, which means room noise, device hum, and any playback quirks end up baked into the file. Conversion can apply to anything, but only the variant that targets audio output includes the demux step.
If you have a video clip with clean recorded audio — a presentation recording, a music performance, a podcast-style clip — extraction is the right word because the audio you want is already inside the file. If you want to capture sound that is playing through your speakers, you need a recording tool instead.
How to Extract Audio from a Local Video File
For a single short clip stored on your computer, a browser-based tool is the fastest path without installing software. The Video to Audio Converter runs entirely in the current tab and keeps the source file on your device throughout the job.
- Open the Video to Audio Converter in a compatible browser and click the file picker. Choose one local video that contains an audio track — MP4, WebM, MOV, M4V, or Ogg up to 500 MiB.
- Click Extract audio and leave the tab open while the tool processes the file. Decoding and recording happen in real time, so the page plays the video from start to finish while it captures the audio track.
- Watch the progress label, which follows playback time. The preview is muted during capture, but the recorded stream still contains the audio from the source clip.
- When the tool reports a finished result, check the listed duration and file size to confirm it matches the source clip.
- Download the Opus WebM audio file. The output is a WebM container with an Opus audio track — not MP3, WAV, AAC, or FLAC.
For longer workflows, or whenever you need lossless output, a different toolset makes more sense. The browser path targets the broadly supported Opus WebM combination; for production work, multichannel audio, or a specific delivery codec, a desktop audio editor with a configurable export is the right choice.
Input Limits That Decide Whether Extraction Starts
Before any extraction begins, the tool checks the file against a set of limits. If any limit fails, the job stops with a visible message rather than producing a broken file. The hard limits are:
| Limit | Value | Why it exists |
|---|---|---|
| Container format | MP4, WebM, MOV, M4V, Ogg | Containers the browser can usually decode |
| File size | Up to 500 MiB | Bounds memory use during decoding |
| Duration | No longer than 5 minutes | Bounds playback time and resource use |
| Longest side | No larger than 4096 pixels | Prevents oversized decoded frames |
| Total resolution | No greater than 3840 × 2160 | Keeps total pixel area within reason |
A filename or MIME type only identifies a possible container; the browser still has to support the codecs inside. That is why a file with a recognized extension can still fail — the codec wrapped inside that container may not be decodable. A video with no audio track fails explicitly instead of producing a silent file.
What the Output File Looks Like
Extraction produces a fresh audio file, not a byte-for-byte copy of the original compressed audio packets. The tool records the browser's decoded audio stream and writes it into a WebM container using the Opus codec at 128 kbps. The audio duration is patched into the WebM Segment Info element so compatible players report a finite timeline instead of an unknown length. The underlying mechanism uses the browser's HTMLMediaElement captureStream and MediaRecorder pipeline, both documented on MDN (HTMLMediaElement captureStream and MediaRecorder).
Because the file is re-encoded, the output's size and quality depend on the Opus encoder rather than the source's original audio codec. A source encoded with a high-bitrate AAC track will not produce an identical-quality Opus file at 128 kbps; Opus is efficient, but it is a different codec with its own compression characteristics. If you need a lossless copy of the original audio, this approach is the wrong tool — extract with a desktop editor that supports remuxing or lossless export.
The format choice is deliberate. WebM with Opus is one of the formats a browser's built-in MediaRecorder can produce without an extra codec dependency, which keeps the tool lightweight. MP3 encoding would require bundling a separate encoder; the tool avoids that by accepting Opus WebM as the output. If you need MP3, AAC, WAV, or FLAC, transcode the resulting Opus WebM with another tool.
Local Browser Extraction vs Upload-Based Services
Two main families of extraction tools exist. The difference between them is where the file goes during processing.
| Aspect | Local browser extraction | Upload-based service |
|---|---|---|
| Where the file is processed | In your current browser tab | On the service's remote server |
| Network needed during work | No upload; file stays on device | Yes, file leaves the device |
| Output formats commonly offered | Whatever the browser can encode (commonly Opus WebM) | MP3, WAV, AAC, FLAC, M4A, OGG, and others |
| File size and duration limits | Bounded by browser memory and decoder | Often set by the service plan |
| Account or sign-up | Usually none | Often required for larger files |
| Privacy posture | Source never leaves device | Source reaches a third-party server |
Local browser extraction is the right fit when the file is short, the output format is acceptable, and the priority is keeping the source on your own device. Upload-based services earn their keep when you need a specific codec, a very large file, or batch processing that a browser cannot handle in one tab.
When Browser Extraction Is the Right Fit
Browser extraction works well for a clear set of tasks:
- Pulling a speech or voiceover out of a presentation recording for a transcript or voice note.
- Capturing a music performance or ambient sound from a short personal clip where Opus WebM is acceptable.
- Isolating a reference track from a screen recording or tutorial clip.
- Producing a small audio file to embed in slides, documents, or a personal archive.
It is not the right fit when you need lossless quality, multichannel preservation, precise trimming with edit points, or a delivery codec that the browser cannot produce directly. Long recordings, master-quality production work, and any task that requires editing the audio rather than just isolating it belong in a dedicated audio editor.
For a deeper walkthrough of the local approach, see the guide on pulling audio from a local video without uploading it.
Use any extraction tool only on content you own or have permission to reuse. Extraction does not bypass DRM, platform access controls, protected streams, remote-only URLs, or copyright restrictions. The technical step of pulling a track out of a container is independent of the legal right to use the resulting audio.