Audio Processing for Internet Radio: How to Make Your Station Sound Professional

by

Play your raw, unprocessed audio stream next to a major commercial radio station and the difference is immediately obvious. Professional stations sound loud, warm, punchy, and consistent. Raw audio sounds flat, thin, and quiet by comparison. The difference is not better music or better equipment — it is audio processing.

Audio processing is the layer between your playout system and your streaming encoder that shapes, controls, and enhances the sound of everything that goes to air. Done right, it transforms a competent-sounding stream into one that holds listeners’ attention. Done wrong, it creates harsh, fatiguing audio that drives people away. This guide will help you get it right.

This guide is part of the Complete Guide to Building a 24/7 Internet Radio Station with PlayoutONE.

What Audio Processing Actually Does

Broadcast audio processing is not a single effect — it is a chain of several processes working together:

  • AGC (Automatic Gain Control): Brings up quiet passages and pulls back loud ones, creating a more consistent average level across tracks and programmes.
  • Multiband compression: Applies compression independently to different frequency ranges (bass, midrange, high frequencies). This prevents loud bass from dominating the mix and keeps each part of the frequency spectrum balanced.
  • Limiting: A hard ceiling that prevents the signal from exceeding a maximum level. Without limiting, loud peaks can cause distortion in the encoder.
  • Stereo enhancement: Widens the stereo image to make the station sound bigger and more spatial on headphones and speakers.
  • Bass enhancement: Adds warmth and body to the low end, which internet audio codecs can sometimes strip away.

These processes work together to produce audio that sounds consistently loud, warm, and controlled — regardless of whether the source track is a quiet acoustic recording or a heavily compressed pop hit.

Where Audio Processing Fits in Your Signal Chain

In a PlayoutONE setup, audio processing sits between your virtual audio cable and your streaming encoder:

PlayoutONE → CABLE Input → [Audio Processor] → LiveStream → Internet

To achieve this, you need a second virtual cable instance (or a second pair of cable devices) if you are using VB-Audio. The routing becomes:

PlayoutONE Output → CABLE Input → Audio Processor Input
Audio Processor Output → CABLE2 Input → LiveStream Input

Alternatively, some audio processors can be inserted as a virtual audio device that sits transparently in the signal path without requiring an additional cable pair. Stereo Tool, for example, can operate as a VST plugin within a compatible host application.

The Right Tools for Internet Radio Processing

Stereo Tool

Stereo Tool is the most widely used audio processor among independent internet radio stations. It delivers broadcast-quality processing in a software application that runs on a standard Windows PC, at a price point — including a generous free tier — that makes it accessible to stations of any size.

The free version includes full processing capability with a watermark tone at intervals. The paid licence (a one-time purchase) removes the watermark and unlocks additional features. For a station going 24/7, the paid licence is worth it.

Stereo Tool is powerful but has a steep learning curve. A large community of presets shared by radio engineers provides excellent starting points — download a few and listen critically before building your own settings from scratch.

Breakaway One

Breakaway One is a dedicated broadcast audio processor with a more streamlined interface than Stereo Tool. If you want solid results without spending hours dialling in parameters, Breakaway One delivers a professional sound with relatively simple configuration. It offers a trial version and is reasonably priced for independent stations.

Omnia and Other Professional Processors

Omnia processors (hardware and software) are used by major commercial broadcasters worldwide. They are exceptionally capable and exceptionally expensive — both in purchase price and in the time required to configure them well. For most independent internet radio stations starting out, Stereo Tool or Breakaway One delivers results that are indistinguishable from Omnia processing to the typical listener.

The Golden Rule: Moderation

Every audio processor gives you the ability to make your station sound louder and more compressed than the previous setting. It is tempting to keep pushing. Do not.

Over-processed audio has a characteristic sound: harsh, fatiguing, and often described as “pumping” or “breathing” — an audible rhythmic in-and-out caused by heavy-handed compression. Listeners experience this as uncomfortable, even if they cannot identify why. They turn it off.

The goal of audio processing for internet radio is controlled consistency, not maximum loudness. A stream that sounds warm, clear, and comfortable at a moderate listening volume will retain listeners far better than one that is the loudest thing in the room for the first thirty seconds and then triggers ear fatigue.

Practical guidance:

  • Start with a conservative preset designed for internet streaming
  • Listen for 30 minutes straight, not just a few seconds
  • Compare against other stations you respect — not against a maximally compressed commercial station
  • Ask someone else to listen and give you honest feedback
  • Make small adjustments and re-evaluate rather than large sweeping changes

Loudness Targets for Internet Streaming

Unlike terrestrial broadcast, internet radio has no regulatory loudness requirements. But there are practical targets that work well:

  • Integrated loudness target: -14 to -16 LUFS — This is where most internet radio stations land. It is loud enough to sound energetic but leaves headroom that keeps the audio clean through the encoder.
  • True peak ceiling: -1 dBTP — Keep your peaks at least 1 dB below digital zero to prevent inter-sample clipping in the encoder.

Most audio processors allow you to set target output levels in LUFS and dBTP. Set these targets in your processor and let the system hold them automatically.

Testing Your Processing

After configuring your processor, test your stream as a listener would hear it:

  1. Open your stream URL in VLC on a different device or in a different room
  2. Listen for at least 20 minutes across different types of music (quiet songs, loud songs, uptempo, downtempo)
  3. Check for: consistent levels, no harsh high frequencies, no pumping or breathing artefacts, no distortion, bass that sounds warm rather than boomy
  4. If you hear any of the above, pull back your processing settings and re-test

What Comes Next