FFmpeg

    FFmpeg is a cross platform tool which allows multimedia files, such as audio recordings, to be converted between different formats. We will give a simple overview on how to use this tool to do basic conversions between audio formats, but keep in mind the features of FFmpeg are far more extensive, and covered in detail within their documentation.

    FFmpeg is a command line tool, which means we need to use a command line interface (CLI), either the command prompt (cmd.exe) or PowerShell for Windows operating systems, or terminal, if you are using macOS or Linux. If this is your first time using FFmpeg, start by downloading and installing the appropriate build from the FFmpeg website. You can then open your CLI, and type ffmpeg, to check that your installation was successful. You should see some information returned, such as the version number of FFmpeg you have installed.

    Installing FFmpeg #

    You can download ffmpeg from its website.

    You can also use a package amanger to install it:

    • Windows
      • winget: winget install ffmpeg
      • Chocolately: choco install ffmpeg
    • MacOS:
      • brew: brew install ffmpeg
    • Linux:
      • apt: apt install ffmpeg
      • yum: yum install ffmpeg

    Converting audio files #

    FFmpeg is able to convert audio easily with smart default settings.

    Simple conversions #

    This is a simple conversion which takes your input file name and format, and returns an output, in the format of the file extension you specify. For example, the code below will take the input (-i) file, named input.mp3, and return a file called output.wav.

    ffmpeg -i input.mp3 output.wav

    Batch conversions #

    We can also convert all files within a directory with a batch conversion. The code below will convert all WAVE files, into MP3. The original file names are preserved for the output, and a _converted suffix is added (which you can modify).

    In PowerShell (Windows, and Linux/Mac if you have PowerShell installed):

    # Converts .m4a files to .wav, for any .m4a files found in the current directory and all sub directories
Get-ChildItem *.m4a -Recurse | ForEach-Object { ffmpeg -i $_ ($_ -replace "\..*$", ".wav") }

# Bonus tip: add -parallel to the for each to convert multiple files at the same time
# (Requires version 7 or above of PowerShell)
Get-ChildItem *.m4a -Recurse | ForEach-Object -parallel { ffmpeg -i $_ ($_ -replace "\..*$", ".wav") }

    In bash (for linux or Mac):

    for i in *.wav; do ffmpeg -i "$i" "${i%.*}.mp3"; done

    Output settings #

    Sample rate #

    FFmpeg will not change sample rate unless you tell it to. If you want to modify the sample rate, first check the audio properties of your sound file by entering the file without specifying an output: ffmpeg -i input.wav. Assuming that input.wav has a sample rate 44100 Hz, but you want to down sample this file, the sample rate of the output file can be specified using -ar. The example below will return a .wav file with a sample rate of 22050 Hz.

    ffmpeg -i input.flac -ar 22050 output.wav

    Bit depth #

    However, FFmpeg defaults to 16-bit encoding when outputting WAVE, and will therefore not preserve bit depth of your audio file, if it is greater than 16. If you want to preserve a 24 bit encoded file, you can manually specify a 24-bit encoder.

    ffmpeg -i input.flac -c:a pcm_s24le output.wav