Hi everyone,

I'm looking for recommendations on the best tools to extract audio from video files. I’ve got some MP4 and MOV files, and I’d like to convert them into high-quality MP3 or WAV files. I’ve tried a few audio extractor from video online tools, but I’m looking for something that retains sound quality and supports batch processing.

Ideally, it would work on both Windows and Mac, and have a user-friendly interface. Any tips on software you've had good experiences with? Free options would be awesome, but I’m open to paid tools if they’re worth it.

Thanks in advance!

 

ffmpeg -i video_file.mp4 -map 0:a:0 -c:a copy output.mp4

Will losslessly extract the first audio stream from a video file. ffmpeg is free & open source and available on all OSs. It's the best tool, but far from the best user experience. I'm sure there's a pretty Mac GUI available though!

 

Put your video on a REAPER track and export the audio to the format of your choosing

 

best way is probably a video editing software like Vegas, Premiere or DaVinci. Or something free open source. Then you can edit the file for your liking before rendering a .Wav or mp3 file

 

If the file has DTS or Dolby, surround or stereo audio, MP4BOX will extract the audio stream along with the video stream into separate files. The audio stream being a .DTS or .AC3 (dolby) file, no losses incurred. At that point it's an easy task to pull that audio stream into any decent audio editor and make your final file. I would suggest .FLAC as a final compressed LOSSLESS format. Last I heard MP4BOX was free.

 

Last edited: Oct 27, 2024 at 8:54 PM

It depends which audio stream you want, because .mp4 or .mov are containers, I don't think you'll find within uncompressed .wav or (lossless) .flac.
Usually are aac, opus or m4a, always compressed.

You can try Virtualdub or Xrecode (supports batch I think) or similar and extract the stream(s), but if you don't like the above formats, you'll have tho recode them (losing quality), and that brings you at the starting point: instead of extract streams, encode clips directly (i.e. mp4 to mp3) with your desired quality (losing quality in anyway).

BTW I suggest before any operation to check audio quality of the video clip with a spectrogram (i.e. spek): if the frequency cut is 15kHz there is no logical sense to convert it in 320kb mp3, and more so in wav or flac.

See below the example: 128kb are enough to retain quality.

 

I would always use a splitter for such a task: e.g. MKVCleaver for MKV files or MP4Box (GUI) for MP4 or MOV files. They can also process multiple files at the same time. They extract the original audio track from the container file. This eliminates the need for additional encoding.

 

Ya right, as you pointed out there will be no mp3 or wav or flac as OP requested, so converting is mandatory in this case.

Spek (or any spectrogram sw) is our friend, as it shows the real bandwidth no matter the audio format/bitrate contained in mp4 or whatever.

 

mkvtoolnix will extract any audio from a video file in a matter of seconds.

 

gMKVextract is a GUI for MKVtoolnix

tsmuxer

Btw: little tools and helpers were the reason, why I had to downgrade to Win10 on the new laptop:
Startmenu with personally organized subfolders and the alias to the program with a short description of what a tool does.
For me impossible to work without, because after a few days I do no longer remember what tool I had needed for what.

Running now W11 in a VM just to find out, if StartAllBack really brings back folders and aliases, because I can't afford to lose my start menu and all the info I have collected over 20 years since WinXP...

 

Really? Isn't the codec, when using lower bandwidth, not raising masking levels throughout the whole spectrum?

 

MKVtoolnix is free for OSX, Linux or Windows. Allows extraction of a single track of video, audio or subs of your choosing
https://mkvtoolnix.download/

 

The whole spectrum bandwidth (MP3) is related to bitrate, having more or less:
128kbps (CBR) cut-off at 16 kHz;
256kbps (CBR) cut-off at 20 kHz;
192kbps (CBR) cut-off at 19 kHz;
320kbps (CBR) cut-off at 20.5 kHz.

So I don't bother masking , the cut-off freq remains.

See the pics below, are the spectrograms from mp4 (barely 14kHz) and then mp3 coded in different increasing bitrates:


Beside the very low encoding of the first mp3 (seen the lofi origin I can barely hear differences, they all suck), all the others are virtually the same and cut-off is the same, no matter the higher bitrate.

You can use choose for any audio to always use 320kb/s, but for poor audio it's like shooting mosquitos with a .600 nitro express, for my purposes I use this simple rule: if cut-off is 16kHz I use 192kb, if higher 320kb, for me it works fine.

I leave the MNR (mask to noise ratio) and all the science behind the band masking to the mp3 algo, I just enjoy to listen.

 

I have been using the "Abelssoft - MusicExtractor" for years, really, fast, easy and clear.
https://musicextractor.software.informer.com/download/#downloading

From the description:

" Extracts the audio track of a video as MP3 With MusicExtractor you can get an MP3 from every video with a simple click and save it as a file. High quality and small file size The sound file has a convincing high quality and a small file size. No unnecessary bells and whistles. One click, done MusicExtractor has no loaded menus and will simply do what you want with one-click. "

 

Last edited: Oct 28, 2024 at 12:52 PM

Thank you @orbitbooster. Great help and even I can understand it!