audio-mixing-patterns

# Audio Mixing Patterns

Comprehensive guide to audio mixing for video production using ffmpeg. Covers narration/music balancing, automatic ducking, timing control, and loudness normalization.

## Core Principle

**Quality Audio = Clear Narration + Supportive Music + Appropriate Levels**

The human voice occupies 85-255 Hz (fundamental) with harmonics up to 8kHz. Music must support, not compete.

## Volume Balancing Formula

```
Standard Video Mix Ratios:
--------------------------
Narration:  100% (reference level)
Music:      15-20% of narration level
SFX:        70-100% of narration level (contextual)

dB Relationships:
-----------------
Narration:  -14 dB LUFS (dialogue standard)
Music bed:  -30 to -35 dB LUFS (under narration)
Music only: -16 dB LUFS (no narration sections)
SFX:        -18 to -20 dB LUFS
```

### Volume Multiplier Quick Reference

| Ratio | Multiplier | Use Case |
|-------|------------|----------|
| 100% | 1.0 | Full volume (narration) |
| 70% | 0.7 | Prominent SFX |
| 50% | 0.5 | Equal blend |
| 30% | 0.3 | Noticeable background |
| 20% | 0.2 | Subtle bed (recommended music) |
| 15% | 0.15 | Minimal presence |
| 10% | 0.1 | Barely audible |

## Basic ffmpeg Mixing Commands

### Two-Track Mix (Narration + Music)

```bash
# Basic mix: narration at full, music at 15%
ffmpeg -i narration.mp3 -i music.mp3 \
  -filter_complex "[0:a]volume=1.0[narr];[1:a]volume=0.15[music];[narr][music]amix=inputs=2:duration=first" \
  -c:a aac -b:a 192k output.m4a
```

### Three-Track Mix (Narration + Music + SFX)

```bash
ffmpeg -i narration.mp3 -i music.mp3 -i sfx.mp3 \
  -filter_complex "\
    [0:a]volume=1.0[narr];\
    [1:a]volume=0.15[music];\
    [2:a]volume=0.7[sfx];\
    [narr][music][sfx]amix=inputs=3:duration=first:weights='3 1 2'" \
  -c:a aac -b:a 192k output.m4a
```

## Timing with adelay Filter

The `adelay` filter positions audio at precise timestamps.

### Syntax

```bash
adelay=delays[|delays...][,all=1]
# delays: milliseconds or samples (with 'S'