FFmpeg Audio Processing Tips: Noise Reduction, Mixing, Volume Adjustment

Overview of FFmpeg Audio Processing

FFmpeg is not only a powerful video processing tool but also equally capable in audio processing. It supports nearly all mainstream audio formats and provides a rich set of audio filters and processing capabilities. From basic format conversion to advanced noise reduction, mixing, and equalization, FFmpeg handles it all efficiently.

This article systematically introduces core FFmpeg audio processing techniques, including noise reduction, multi-track mixing, volume normalization, fade in/out, equalizer adjustment, and other practical features to help you improve audio processing efficiency and quality.

Audio Noise Reduction

Audio noise reduction is one of the most common requirements in audio processing. FFmpeg provides various noise reduction filters suitable for different scenarios.

afftdn Frequency Domain Denoising

afftdn is a frequency-domain based noise reduction filter with excellent results and adjustable parameters:

# Basic noise reduction
ffmpeg -i input.wav -af "afftdn" output.wav

# Specify noise reduction strength (-30dB to 0dB)
ffmpeg -i input.wav -af "afftdn=nf=-25" output.wav

arnndn AI Denoising

arnndn uses recurrent neural networks for noise reduction, requiring model file support for more intelligent results:

ffmpeg -i input.wav -af "arnndn=m=model.rnnn" output.wav

highpass/lowpass Simple Denoising

For simple low-frequency or high-frequency noise, you can use high-pass and low-pass filters:

# Remove low-frequency noise (below 80Hz)
ffmpeg -i input.wav -af "highpass=f=80" output.wav

# Remove high-frequency noise (above 10kHz)
ffmpeg -i input.wav -af "lowpass=f=10000" output.wav

Multi-Track Audio Mixing

Mixing multiple audio tracks together is a common operation in audio production.

Two-Channel Audio Mixing

Use the amix filter to mix multiple audio inputs:

ffmpeg -i voice.wav -i bgm.wav -filter_complex "amix=inputs=2:duration=shortest" output.wav

Adjusting Track Volume Ratios

Use the volume filter to adjust each track's volume separately before mixing:

ffmpeg -i voice.wav -i bgm.wav -filter_complex \
  "[0:a]volume=1.0[a1];[1:a]volume=0.3[a2];[a1][a2]amix=inputs=2:duration=first" \
  output.wav

Audio Sequencing (One After Another)

Use the concat filter to concatenate audio in sequence:

ffmpeg -i intro.wav -i main.wav -i outro.wav -filter_complex \
  "[0:a][1:a][2:a]concat=n=3:v=0:a=1" \
  output.wav

Volume Adjustment and Normalization

Adjusting audio volume is the most basic and commonly used audio processing operation.

Basic Volume Adjustment

# Increase volume by 50%
ffmpeg -i input.wav -af "volume=1.5" output.wav

# Reduce to half
ffmpeg -i input.wav -af "volume=0.5" output.wav

# Increase by 3dB
ffmpeg -i input.wav -af "volume=3dB" output.wav

Volume Normalization

The loudnorm filter can normalize audio to a specified loudness level, an essential tool for professional audio processing:

# EBU R128 normalization (-16 LUFS)
ffmpeg -i input.wav -af "loudnorm=I=-16:TP=-1.5:LRA=11" output.wav

Automatic Gain Control

The dynaudnorm filter can dynamically adjust volume to keep overall loudness consistent:

ffmpeg -i input.wav -af "dynaudnorm=f=200:g=15" output.wav

Fade In/Out Effects

Fade in/out effects make audio beginnings and endings more natural.

Basic Fade In/Out

# 3-second fade in at the beginning
ffmpeg -i input.wav -af "afade=t=in:st=0:d=3" output.wav

# 3-second fade out at the end
ffmpeg -i input.wav -af "afade=t=out:st=10:d=3" output.wav

Combined Fade In/Out

Add both fade in and fade out effects simultaneously:

# 2-second fade in, 2-second fade out (15-second total duration)
ffmpeg -i input.wav -af "afade=t=in:st=0:d=2,afade=t=out:st=13:d=2" output.wav

Equalizer Adjustment

FFmpeg provides various equalizer filters to adjust audio frequency response.

Basic Equalizer

Use the equalizer filter to adjust gain at specific frequencies:

# Boost 1kHz midrange by 3dB
ffmpeg -i input.wav -af "equalizer=f=1000:t=q:w=1:g=3" output.wav

# Cut 100Hz bass by 5dB
ffmpeg -i input.wav -af "equalizer=f=100:t=q:w=2:g=-5" output.wav

Multi-Band Equalizer

Combine multiple equalizers for multi-band equalization:

ffmpeg -i input.wav -af \
  "equalizer=f=80:t=q:w=1:g=-2,equalizer=f=250:t=q:w=1:g=1,equalizer=f=1000:t=q:w=1:g=2,equalizer=f=4000:t=q:w=1:g=1,equalizer=f=8000:t=q:w=1:g=-1" \
  output.wav

Audio Compression

Audio compressors reduce dynamic range for more consistent volume, commonly used in broadcasting and podcast production.

Basic Compression

# Basic compression effect
ffmpeg -i input.wav -af "acompressor" output.wav

# Custom parameters
ffmpeg -i input.wav -af "acompressor=threshold=-20dB:ratio=4:attack=5:release=50" output.wav

Limiter

The alimiter limiter prevents audio overload and clipping:

ffmpeg -i input.wav -af "alimiter=limit=-1dB" output.wav

Audio Reverse

Audio reverse is a special effect sometimes used in creative audio production.

# Reverse the entire audio
ffmpeg -i input.wav -af areverse output_reverse.wav

Other Practical Tips

Extract Audio from Video

# Extract as MP3
ffmpeg -i input.mp4 -vn -c:a libmp3lame -q:a 2 output.mp3

# Extract as WAV
ffmpeg -i input.mp4 -vn -c:a pcm_s16le output.wav

Audio Format Conversion

# WAV to MP3
ffmpeg -i input.wav -c:a libmp3lame -b:a 320k output.mp3

# MP3 to AAC
ffmpeg -i input.mp3 -c:a aac -b:a 256k output.aac

Summary

FFmpeg's audio processing capabilities are extremely powerful — from basic volume adjustment and format conversion to advanced noise reduction, mixing, and equalization, all can be done efficiently via the command line. Mastering these techniques can significantly improve audio processing efficiency and quality.

Whether it's podcast production, video dubbing, or music post-production, FFmpeg is an audio processing tool worth learning in depth.