HEIC to JPG: Quality, Metadata, and Privacy
A careful HEIC-to-JPG workflow preserves the picture you need without accidentally sharing location data or repeatedly degrading the image.
Read guide →Extract Audio
Save the audio track from a video as a variable bitrate MP3, encoded on your own device with the LAME encoder.
Drop a video here or click to browse
MP4, WebM, MOV, AVI — processed locally and never uploaded.
Reviewed 2026-08-15
Free Extract Audio – how it works
The video stream is discarded and the audio is re-encoded to MP3 with libmp3lame in variable bitrate mode. VBR spends more bits on complex passages and fewer on quiet ones, which gives a better result per megabyte than a fixed bitrate at the same average size.
Three quality settings map to LAME's VBR scale, where lower numbers mean higher quality. Setting 0 is roughly 245 kbps average and is close to transparent for music. Setting 2 sits around 190 kbps and is the usual recommendation. Setting 5 is near 130 kbps and is fine for speech.
Before the audio stage runs, the file is probed to confirm it actually has a sound track. A video with no audio produces a clear message rather than a zero length MP3.
How to use Extract Audio online
- Add the video. Drop in the clip you want the sound from. Any container FFmpeg reads will work, including MP4, WebM, MOV, AVI, and MKV.
- Choose MP3 quality. Use best for music you care about, high for general use, and medium for speech, lectures, or interviews where file size matters more than fidelity.
- Extract and download. Run the extraction and save the MP3. Audio-only encoding is much faster than video work, so this finishes quickly even on longer files.
When to use it
- Keep the audio from a recorded talk or interview for transcription.
- Save a soundtrack from a video you shot yourself.
- Turn a recorded meeting into an MP3 you can listen to while walking.
- Pull a voiceover out of a rough cut so it can be re-used against new footage.
Known limitations
- MP3 is a lossy format, so encoding from a video's already compressed audio loses a little more. For archival work, a lossless container is a better destination.
- Only MP3 output is offered here. There is no WAV, FLAC, or AAC target.
- Videos with several audio tracks, such as multiple language dubs, will give you the default track rather than a choice.
Frequently asked questions
What audio quality should I pick?
Use high, the VBR 2 setting, unless you have a reason not to. It averages around 190 kbps and is hard to distinguish from the source for most listeners. Drop to medium for speech-only recordings.
Does this work if the video has no sound?
The file is checked for an audio stream first. If there is none, you get an explicit message instead of an empty MP3 file.
Is my video uploaded to a server?
No. FFmpeg is compiled to WebAssembly and runs inside the page, so the file is read from your disk into the tab's own memory. Nothing about the video is sent to opentools.
Can I get lossless audio out instead?
Not with this tool. The output is always MP3. Extracting the original audio stream without re-encoding needs a desktop FFmpeg build with a stream copy to a matching container.
Why does the first render take longer?
The FFmpeg WebAssembly build is about 31 MB and has to download before the first job can start. The browser caches it afterwards, so later visits skip that step and begin encoding almost immediately.
Related free tools
Browse every video tools, or jump straight to extract.
Guides related to Extract Audio
PNG vs JPG vs WebP vs AVIF: Which Format Should You Use?
There is no universally best image format: photographs, interface graphics, archives, and web pages have different requirements.
Read guide →How to Compress Images for the Web Without Visible Quality Loss
The best web-image optimisation combines right-sized pixels with a suitable format and a quality setting tested on real content.
Read guide →