How to Use FFmpeg Bandpass Filter to Isolate Frequencies

This article provides a quick guide on how to use the bandpass audio filter in FFmpeg to isolate a specific frequency range. You will learn the basic syntax of the filter, understand its key parameters like center frequency and band width, and see practical command-line examples to apply to your audio files.

The bandpass filter in FFmpeg allows a specific band of frequencies to pass through while attenuating (quieting) frequencies below and above this range. It is highly useful for tasks like isolating human speech, removing low-end rumble and high-end hiss simultaneously, or targeting specific audio elements.

Basic Syntax

The basic syntax for applying the bandpass filter in FFmpeg is:

ffmpeg -i input.mp3 -af "bandpass=f=CENTER_FREQUENCY:w=BAND_WIDTH" output.mp3

Key Parameters

• f (frequency): The center frequency of the band you want to keep, specified in Hz. The default value is 3000 Hz.
• width_type (optional): Defines how the band width is measured. Common options include:
  • h (Hz - default)
  • q (Q-factor)
  • o (octaves)
• w (width): The width of the filter band. The default value is 200 Hz.

Practical Example: Isolating Mid-Range Frequencies

To isolate a frequency range centered at 1500 Hz with a width of 1000 Hz (allowing frequencies roughly between 1000 Hz and 2000 Hz to pass through clearly), use the following command:

ffmpeg -i input.wav -af "bandpass=f=1500:w=1000" output.wav

Using Q-Factor for Narrower Isolation

If you want a sharper, more precise cut-off around your target frequency, you can change the width_type to Q-factor (q). A higher Q-factor results in a narrower band.

To isolate a narrow band around 1000 Hz using a Q-factor of 2.0, run:

ffmpeg -i input.wav -af "bandpass=f=1000:width_type=q:w=2.0" output.wav