How to Use the FFmpeg tiltsdr Audio Filter
Spectral tilt alters the overall frequency balance of a sound
recording across the audible spectrum, creating a warmer or brighter
tonal character by applying a continuous gain slope. This guide explains
how to use FFmpeg's tiltsdr filter to apply a spectral tilt
to an audio file, covering command-line syntax, key configuration
parameters, and practical usage examples.
Basic Syntax
The tiltsdr filter is applied using FFmpeg’s audio
filter graph flag (-af). The basic command structure
is:
ffmpeg -i input.wav -af "tiltsdr=option1=value1:option2=value2" output.wavIf no parameters are passed, the filter runs using its default parameters, applying standard linear spectral attenuation or boost depending on the input configuration.
Key Filter Parameters
The tiltsdr filter accepts parameters that dictate how
and where the spectral slope is applied across the frequency range:
- tilt / t: Sets the slope of the spectral tilt in decibels (typically per octave or across the entire frequency range). Positive values tilt the spectrum upward (boosting high frequencies and reducing low frequencies), while negative values tilt downward (boosting lows and attenuating highs).
- frequency / f: Defines the pivot or center frequency in Hertz (Hz). Frequencies above this point are boosted or cut in opposition to frequencies below this point. The default is typically set to the acoustic midrange, such as 1000 Hz.
- gain / g: Sets an overall gain offset applied across the processed output to prevent clipping caused by excessive high-frequency or low-frequency boosts.
Practical Examples
1. Brightening Dark Audio
To lift the high frequencies and simultaneously roll off heavy low end, apply a positive tilt value with a pivot around 1000 Hz:
ffmpeg -i input.mp3 -af "tiltsdr=tilt=2.0:frequency=1000" -c:a pcm_s16le output.wav2. Creating a Warmer, Darker Sound
To dampen harsh high frequencies and add warmth across the lower registers, apply a negative tilt value:
ffmpeg -i input.flac -af "tiltsdr=tilt=-1.5:frequency=1000" output.flac3. Preventing Digital Clipping with Gain Adjustment
Significant upward tilts can push peak signals beyond 0 dBFS. Incorporate the gain parameter to compensate for high-end boost:
ffmpeg -i voice.wav -af "tiltsdr=tilt=3.0:frequency=1500:gain=-3.0" voice_tilted.wavIntegrating with Video Files
If you are modifying audio within a video container without re-encoding the video track, combine the filter with video stream copying:
ffmpeg -i input.mp4 -c:v copy -af "tiltsdr=tilt=1.0:frequency=1000" -c:a aac -b:a 192k output.mp4This updates the spectral profile of the audio track while preserving the original video quality and saving processing time.