Configure FFmpeg Chorus Filter Delays Feedback and Speed

The FFmpeg chorus filter is a powerful tool for adding depth and texture to audio files by simulating multiple voices or instruments playing together. This article provides a straightforward guide on how to configure the key parameters of the chorus filter—specifically delays, feedback (decay), speeds, and depths—enabling you to customize the intensity and character of the chorus effect in your audio processing workflows.

Understanding the Chorus Filter Syntax

The basic syntax for the FFmpeg chorus filter is:

chorus=in_gain:out_gain:delays:decays:speeds:depths

When adding multiple voices to create the chorus effect, you specify multiple values for delays, decays, speeds, and depths. Each voice's parameters are separated by a pipe (|) symbol.

1. Configuring Delays

The delays parameter specifies the delay time in milliseconds for each voice.

  • Values: Typically between 20ms and 50ms.
  • How it works: Shorter delays (around 20ms) create a tighter, more subtle double-tracking effect. Longer delays (above 40ms) create a more distinct, spacious separation between the original signal and the chorus voices.
  • Syntax Example: For a three-voice chorus, you might use 40|50|60 to space out the entry of each voice.

2. Configuring Decays (Feedback/Volume)

The decays parameter controls the volume or attenuation of each delayed voice relative to the input signal. It acts as the feedback or mix level for each voice.

  • Values: Decimal numbers between 0.0 (silent) and 1.0 (maximum volume). Normal range is 0.2 to 0.4.
  • How it works: Higher decay values make the chorus effect much more pronounced and louder, while lower values keep the effect subtle and blended in the background.
  • Syntax Example: 0.3|0.25|0.3 sets slightly different volumes for three voices to sound more natural.

3. Configuring Speeds

The speeds parameter determines the modulation speed of the Low-Frequency Oscillator (LFO) in Hertz (Hz) for each voice.

  • Values: Typically between 0.1 and 2.0 Hz.
  • How it works: Speed controls how fast the pitch of the chorus voice fluctuates. Lower speeds (e.g., 0.2 to 0.5 Hz) create a gentle, slow sweeping effect. Higher speeds (above 1.5 Hz) create a rapid vibrato or warbling effect.
  • Syntax Example: 0.25|0.4|0.3 modulates each voice at a slightly different rate to prevent them from modulating in unison, which sounds more organic.

4. Configuring Depths

The depths parameter dictates the LFO modulation depth in milliseconds. It determines how far the delay time deviates from the set delays value.

  • Values: Typically between 1.0 and 5.0 ms.
  • How it works: Higher depths result in wider pitch variations, making the detuning effect of the chorus much more obvious. Lower depths keep the pitch modulation tight and subtle.
  • Syntax Example: 2.0|2.5|1.5

Practical Command Examples

Subtle, Natural Chorus (Two Voices): This configuration uses moderate delays and slow speeds to add a gentle warmth to the audio.

ffmpeg -i input.mp3 -af "chorus=0.7:0.9:40|50:0.3|0.25:0.25|0.4:2.0|2.5" output.mp3

Heavy, Sweeping Chorus (Three Voices): This configuration uses longer delays, higher decay levels, and distinct modulation speeds to create a thick, dramatic effect.

ffmpeg -i input.mp3 -af "chorus=0.5:0.9:50|60|70:0.4|0.35|0.3:0.5|0.7|0.9:3.0|3.5|4.0" output.mp3