CloudRadio automatically processes every track you upload to your media library. This article explains what happens and why.
What CloudRadio does to your audio
When you upload a track, CloudRadio runs a processing pipeline that handles loudness normalization, metadata extraction, waveform generation, and format compatibility. All of this happens in the background. You can start using your track as soon as processing completes.
Loudness normalization
The most important processing step is loudness normalization. Without it, different tracks play at different volumes, forcing listeners to constantly adjust their volume control.
CloudRadio normalizes every track to a consistent loudness target:
| Parameter | Target | What it means |
|---|---|---|
| Integrated loudness | -14 LUFS | The overall perceived loudness of the track |
| True peak | -1 dBTP | The maximum signal level, with headroom to prevent clipping |
| Loudness range | 11 LU | How much the loudness varies within the track |
LUFS (Loudness Units Full Scale) is the broadcast industry standard for measuring perceived loudness. Unlike simple peak or RMS measurements, LUFS accounts for how human hearing works, weighting frequencies the ear is most sensitive to.
The -14 LUFS target is slightly louder than the EBU R128 broadcast standard (-23 LUFS) and aligns with what major streaming platforms use. This keeps your station sounding full and present without over-compressing the audio.
How normalization works
CloudRadio uses a two-pass process:
- Measurement pass: Analyzes the entire track to determine its current loudness, peak level, and loudness range.
- Normalization pass: Applies precise gain adjustments using the measured values. This produces accurate, transparent results.
For MP3 files, CloudRadio uses lossless gain adjustment when possible. This modifies the MP3 header data without decoding and re-encoding the audio, preserving the original sound quality exactly.
If lossless adjustment isn’t compatible with a particular MP3 file, CloudRadio falls back to the two-pass normalization method.
Cue point detection
CloudRadio analyzes the beginning and end of each track to detect silence. It sets suggested cue-in and cue-out points that skip leading and trailing silence. This keeps transitions between tracks tight and professional.
You can override these cue points manually in the media library if needed.
Waveform generation
CloudRadio generates a visual waveform for every track. This appears in the media library and playlist editors, giving you a visual reference of the track’s dynamics. The waveform also helps you verify cue points at a glance.
Artwork extraction
If your audio file contains embedded artwork (common in MP3 and M4A files), CloudRadio extracts it and displays it in the media library, now-playing widgets, and player embeds.
If no embedded artwork is found, CloudRadio attempts to find a match using the track’s artist and title.
Format handling
CloudRadio accepts a range of audio formats, including both CBR and VBR encoded files. How each is processed depends on the source format:
| Source format | What happens |
|---|---|
| MP3 | Stored as-is. Lossless loudness adjustment applied. No re-encoding. |
| OGG Vorbis | Stored as-is. No re-encoding. |
| M4A/MP4 (AAC) | Remuxed for streaming compatibility. Audio codec preserved. No re-encoding. |
| WAV | Transcoded to MP3 at a bitrate matched to source quality. |
| FLAC | Transcoded to MP3 at a bitrate matched to source quality. |
| AIFF | Transcoded to MP3 at a bitrate matched to source quality. |
| ALAC (Apple Lossless) | Transcoded to MP3 at a bitrate matched to source quality. |
When transcoding lossless formats, CloudRadio selects a bitrate based on the source file’s characteristics. It uses standard CBR tiers (96, 128, 160, 192, 256, or 320 kbps) and never upscales beyond what the source quality justifies. See Bitrates and Codecs for details on why.
Loudness for broadcasters
If you broadcast live or use external automation software, your audio bypasses the media library processing pipeline. In that case, you’re responsible for your own loudness management.
Tips for consistent live audio
- Use a limiter in your broadcasting software chain. Set the ceiling to -1 dBTP to prevent clipping.
- Normalize your music library in your DJ or automation software before broadcast. Most DJ software includes loudness analysis (often called “auto-gain” or “track normalization”).
- Target -14 LUFS if your software supports LUFS metering. This matches what CloudRadio applies to media library tracks, so the transition between AutoDJ and live will sound seamless.
- Avoid over-compression. Heavy compression makes audio sound loud but fatiguing over time. Aim for natural dynamics with gentle limiting on peaks.
Recommended processing chain for live broadcasting
- EQ (optional): Remove low-frequency rumble below 80 Hz, add subtle presence around 3-5 kHz for voice clarity.
- Compressor: Gentle ratio (2:1 to 4:1), moderate attack and release. Tames peaks without squashing dynamics.
- Limiter: Brick-wall limiter at -1 dBTP. Catches anything the compressor misses.
- Output level: Keep your encoder input at a healthy level without clipping. Most encoders show a meter. Stay out of the red.
Related
- Media Library for uploading and managing your tracks
- Bitrates and Codecs for choosing the right codec and bitrate
- Stream Settings for configuring your encoder output
- What Is HLS Streaming? for how HLS delivers audio to listeners