Video to Audio

Most video files already carry their sound as a separate stream, so taking it out is quick and does not touch the picture at all. The reasons to do it vary: a phone recording of a concert you want on your playlist, a recorded lecture you want to listen to on the bus, an interview you need to edit or transcribe, or a clip whose music you want to use in another project. Each of these points to a different format and different settings, so the cards below group the common cases.

Batches help when the same job repeats. Drop a folder's worth of recorded classes or a season of video podcasts at once, up to 20 files, and every file gets the same format and settings. If some need different settings, run them as a second batch.

Nearly every video stores its sound already compressed, typically as AAC in MP4 and MOV files or Opus and Vorbis in WebM. When you extract it, the track is decoded and then written in the format you pick. Writing to WAV, FLAC or AIFF stores that decoded sound bit for bit, so nothing more is lost; the file gets larger because those formats hold far more data per second than the compressed track did. Writing to MP3, M4A, AAC, OGG, Opus or WMA encodes it again, which keeps files small at the cost of a second, usually small, loss.

Which way to go depends on what happens next. If the audio is the final product, for listening or sharing, a compressed format is the sensible choice and MP3 is the one that opens everywhere. If you will edit, mix or master it, start from a lossless file so every later save works from the cleanest copy. The sheet below summarizes MP3, the most requested output; MDN's audio codec guide covers the others.

MP3 is the one format every phone, car, speaker and app plays. M4A and AAC hold the same AAC codec; M4A wraps it in an MP4 box that also carries tags and cover art, while AAC is the bare stream that some hardware players ask for. OGG usually means Vorbis audio and is common in games and on Linux. Opus, standardized by the IETF as RFC 6716, was designed for voice and music over the internet and holds up well at low bitrates. WMA is only worth picking for older Windows software that requires it.

On the lossless side, WAV is the plain, uncompressed format that every editor opens, and AIFF is its counterpart from the Apple world. FLAC stores the same sound in a smaller file by packing it without loss, and it is now documented by the IETF as RFC 9639. If you are unsure, pick MP3 for listening and WAV for editing; you can always convert again later from the lossless copy.

Bitrate appears for the compressed formats and runs from 64 to 320 kbps, with 192 kbps as the default. That is a comfortable level for music. Speech sounds clear well below it, so talks and voice notes can go down to 128 kbps or lower. Because the video's own sound was compressed when it was recorded, raising the bitrate far above the original rarely makes a difference you can hear.

Sample rate stays at Same as source unless a device or program asks for something specific. Video sound is usually recorded at 48 kHz, CDs use 44.1 kHz, and many speech recognition tools work at 16 kHz. Channels can turn stereo into mono, which halves the size of lossless files and loses nothing when only one person is talking. Bit depth applies to WAV, FLAC and AIFF; 16-bit is CD quality and 24-bit leaves headroom for editing.

Normalize volume brings quiet recordings up to a standard loudness, so a talk filmed from the back of a room becomes easier to hear. Audio track picks between several sound tracks in one file, which films, dubbed shows and screen recordings with a separate microphone sometimes have. Track 1 is the main one.

Start at and End at under Trim keep only a section, in seconds or minutes:seconds. That saves a separate cutting step when you want a single song from a concert or one answer from a long interview. For finer cuts with fades, extract the whole track first and then use the Trim Audio tool.

The extracted file is silent or too quiet. Check that the video really has sound on the track you chose; some recordings put the microphone on track 2. If the level is simply low, run it again with Normalize volume on.

The audio is out of step when added back to the video. Extracting does not change the timing, but cutting with Start at does. Use the same start time in your editor, or extract the whole track and line it up there.

The file is much larger than the video's sound. That is expected for WAV, FLAC and AIFF, which hold uncompressed or losslessly packed audio. Choose MP3 or M4A when size matters more than editing headroom.

An old car stereo or speaker does not play the file. Many older devices read only MP3, often at 44.1 kHz. Pick MP3, set Sample rate to 44.1 kHz, and keep the bitrate at 192 kbps or below.

If you already have an audio file and only need a different format, Audio Converter does that with the same settings, and the most common video case, an MP4 file to MP3, has its own page at MP4 to MP3.

Extracting audio is a technical step and works the same on any file, but the right to reuse the result depends on who made the video. Your own recordings, files you have a license for and material released for reuse are fine. Music videos, films and broadcasts are usually protected, and their sound may not be shared or published without permission even after it has been saved as a separate file. When in doubt, check the license that came with the video or ask the person who made it.