Extract Audio from Video
Save the audio track of a video as a 192 kbit/s MP3. A file with no audio track is refused, not silently emptied.
What this tool extracts
One audio track, as a 192 kbit/s MP3. The video stream is dropped rather than copied, so the result is a few megabytes of sound instead of a container that still carries the picture. MP3 is deliberate: it plays everywhere, including old car stereos and presentation software that refuses an AAC file.
The upload is probed first. A video with no audio track is refused with a sentence that says so, because returning a zero-length file would look like a success.
How to use it
Choose a video file and press Run. The page returns the MP3 plus the duration it read from the container, so you can check that the whole clip was covered.
Limits
20 MB of input, 2 minutes of duration and 24 MB of output, with a 180-second cap on the encode. A file outside those limits is refused before the encoder starts.
Honest about the audio
The source track is re-encoded to MP3 at 192 kbit/s, which is transparent for speech and close to it for most music, but it is a second encode, not the original stream. If you need the untouched audio, remux the file locally with a stream copy instead.
Upload a video and this page returns just its audio track, encoded as a 192 kbit/s MP3 named `extract-audio-from-video.mp3`, with the duration and the produced size in the summary.
It takes the first audio stream and discards the picture entirely, so a video with several audio tracks gives you the first one. A file with no audio track is refused with that reason rather than returning a silent file.
The limits are the tool's own: 20 MB in, two minutes of video, and 24 MB of produced audio. The upload is sent to the tool server for the run and is not stored.
What is kept and what is re-encoded
The audio is decoded and re-encoded to MP3 at a fixed bitrate, so the result is a fresh file rather than the original audio stream copied out. Its size follows the bitrate and the duration, not the source container.
The produced file is inspected before it is returned and must actually carry an audio stream, so a conversion that silently produced nothing is reported as a failure.
A two-minute ceiling is measured on the input, which is why a long screen recording is refused before any encoding starts.
- Output: MP3, 192 kbit/s
- First audio stream only
- Input ceiling: 20 MB and 2 minutes
When the MP3 is not what you want
MP3 is lossy. If the source audio was already compressed, extracting it re-encodes a second time and loses a little more. A tool that copies the original stream would avoid that, and this page does not.
For a podcast or a music file, the original audio track inside the container is usually better than an extraction. Use this page when you need a standalone file, not when you need the best quality copy.
How to use the video upload
Upload the video, within 20 MB and two minutes, and run the tool. The first audio stream is taken and the picture discarded, so a multi-track file yields its first track.
Download `extract-audio-from-video.mp3` at 192 kbit/s with duration and size in the summary. A file with no audio track is refused with that reason rather than returning silence.
What to Expect
Best for
- Pulling the sound out of a short screen recording
- Getting a standalone MP3 of a two-minute clip
- Checking whether a video carries an audio track at all
Not ideal for
- Long recordings, which exceed the two-minute ceiling
- Lossless extraction, since the output is a re-encoded MP3
- Choosing between several audio tracks, of which only the first is used
What this tool keeps
- The full duration of the first audio stream
- A named output file and a reported size
- A refusal, not a silent file, when there is no audio
What may need cleanup
- A second generation of lossy encoding
- The missing second or third audio track of a multi-language file
- The fixed 192 kbit/s bitrate, which may be higher than the source
Common errors
- Uploading a video without sound and reading the refusal as a failure
- Expecting the original codec to be copied through
- Trying a three-minute clip and hitting the duration limit
Example use cases
- Extracting a voice memo from a screen recording
- Getting the audio of a short clip for a transcript
- Checking whether a submitted video has any sound
Sample input
A 45-second screen recording with a voice-over.
Sample output
An MP3 of the same 45 seconds at 192 kbit/s, with its size reported alongside the duration.
Who this is for
- People who need the sound from a short clip
- Anyone transcribing a recording
- Teams checking a submitted video for audio
Frequently asked questions
What audio file does the extractor return?
A 192 kbit/s MP3 named extract-audio-from-video.mp3, with the duration and produced size in the summary.
Is the extraction lossless?
No. The output is a re-encoded MP3, so a second generation of lossy encoding is possible.
What if the video has several audio tracks?
Only the first audio stream is used, so a second language track is not extracted.
Build custom tools for your store
Join the SDK program to create useful ecommerce utilities for your store.