MELD Master

Measured, mastered and downloaded on your own machine. Nothing is uploaded.

Drop a track

WAV, FLAC, MP3 or AIFF. Everything runs on this machine — nothing is uploaded.

Source

Loudness—LUFS

True peak—dBTP

Dynamic range—dB PLR

Target

Match a reference

nothing is uploaded

Drop in a song whose tone you want, and MELD moves your track toward it. Both files stay on your machine — nothing is uploaded.

After mastering — A/B at matched loudness

Achieved—LUFS

True peak—dBTP

Dynamic range—dB PLR

Gain applied—dB

Max limiting—dB

Rendered in—sec

Limiter—

Levels matched — you are hearing the processing, not the volume.

Everything — measurement, mastering and the download — runs on your own machine. Your audio never leaves it, except a track you choose to post.

What it does

Drop a track or an episode in and MELD Master measures it, renders it to your chosen target, and lets you hear the two versions back to back at matched loudness — so you are judging the work, not just the volume. It runs entirely in your browser. Nothing is uploaded, nothing is stored, and it keeps working if your connection drops after the page has loaded.

Measure

ITU‑R BS.1770 gated integrated loudness and 4× oversampled true peak, read before anything is processed.

High-pass

30 Hz, to clear sub-rumble that eats headroom you cannot hear.

Level

Gain toward your chosen target: −14 LUFS for Music, Spotify Podcasts and YouTube, −16 for Apple Podcasts, or any figure you set yourself. Only Spotify and Apple publish a figure at all — which of these are official and which are convention is worth two minutes.

Limit

True-peak limiter with 5 ms lookahead, a −1 dBTP ceiling and adaptive release.

Verify

The render is measured again afterwards, and you are told what it actually achieved — not what it aimed for.

Export

24-bit WAV with the achieved loudness and true-peak figures written into the file metadata.

How to use it

Drop your track onto the panel above, or choose a file. WAV, FLAC, MP3 and AIFF all work.
Read the source figures. Loudness, true peak and dynamic range are measured before any processing.
Pick a target. Music is the sensible default if you are unsure. For spoken word, Apple Podcasts at −16 is the safer delivery level.
Render it, then A/B the original against the result. They are matched for loudness, so what you hear is the processing rather than the level.
Download the WAV. The achieved figures are written into the file metadata.

A word of honesty: this will not rescue a mix that is already clipped or crushed. If the source has no dynamic range left, MELD Master says so rather than pretending. Master from the best mix you have.