# Audio Restoration

> Master audio restoration techniques including noise reduction, de-clicking, de-humming, and spectral repair for cleaning imperfect recordings.

- Skill: `majiayu000/audio-restoration` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/audio-restoration`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/audio-restoration/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/audio-restoration

---


# Audio Restoration

Professional audio restoration techniques.

## Overview

Audio restoration involves using software techniques to identify and remove imperfections in recorded audio, addressing issues from background noise to clicks, hums, and distortion. This process is crucial in music production, post-production, and content creation to improve audio quality and avoid re-recording.

This skill covers core restoration techniques including spectral editing, noise print capture, adaptive noise reduction, de-clicking, de-humming, and the use of industry-standard software like iZotope RX, Acon Digital Acoustica, CEDAR Studio, and Adobe Audition.

## Core Concepts

### Core Restoration Techniques

**Visual Identification** — Use spectral frequency displays to pinpoint and select specific noise ranges or individual artifacts

**Noise Print Capture** — "Learn" noise characteristics by analyzing segment containing only noise; creates profile for removal throughout recording

**Adaptive Noise Reduction** — Real-time reduction of crackle, wind rumble, tape hiss, power-line hum

**DeClipper** — Remove clipping distortion

**DeHummer** — Remove electrical hum and harmonics

**Spectral Editing** — Manipulate complex areas in spectrogram like image editing software

### Key Parameters

**Noise Reduction Graph** — Frequency (x-axis) vs. reduction amount (y-axis)

**Noise Floor** — Highest and lowest amplitude of detected noise at each frequency

**Reduction Amount** — Percentage of noise reduction; typically 5-20 dB for high quality

**Advanced Settings** — Spectral Decay Rate, Smoothing, Precision Factor, FFT Size

## Professional Workflows

### Restoration Workflow

1. **Import and Assess** — Load audio file, identify problems through listening and visual analysis
2. **Capture Noise Print** — Select section with only noise, create noise profile
3. **Apply Noise Reduction** — Use learned profile to reduce noise throughout recording
4. **De-Click/De-Crackle** — Remove clicks, pops, and vinyl crackle
5. **De-Hum** — Remove 50/60 Hz hum and harmonics
6. **Spectral Repair** — Fix specific artifacts using spectral editing
7. **De-Clip** — Restore clipped audio
8. **Final Quality Check** — Listen critically, verify no artifacts introduced
9. **Export** — Save restored audio in appropriate format

### Software Tools

**iZotope RX** — Industry standard with AI-powered modules, spectral editor, comprehensive repair tools

**Acon Digital Acoustica** — Cost-effective alternative with deep learning dialogue extraction

**CEDAR Studio** — Original innovator with DNS dialogue noise suppression

**Adobe Audition** — Comprehensive suite with adaptive noise reduction and spectral display

## Tools and Software

Professional software and equipment used in the industry for audio restoration.

## Best Practices

Guidelines and recommendations for achieving professional-quality results in audio restoration.

## Common Challenges

Typical issues encountered in audio restoration and their solutions.

## Using the Reference Files

**`/references/fundamentals.md`** — Core concepts and foundational techniques.

**`/references/advanced-techniques.md`** — Professional workflows and advanced methods.

**`/references/workflows-best-practices.md`** — Industry standards and optimization.

**`/references/troubleshooting.md`** — Common problems and solutions.