A finished render in AudioColor is not one file. It is a small bundle, and each piece exists because some part of a real edit needs it. Here is what to do with each one.
Start with the timeline, not the audio
If you are finishing in Final Cut Pro or DaVinci Resolve, export the FCPXML and import that first. It carries the clip order, the in and out points, and the names — so the moment it lands, your cuts are already laid out. The numbered WAVs sit next to it and get linked automatically.
For Premiere Pro, use the XML export instead. It is XMEML, which is the format Premiere has understood for years, and it does the same job: the timeline arrives assembled rather than as a folder of clips you have to place by hand.
When to use the numbered WAVs on their own
Reach for the raw clips when you are not cutting to a timeline at all:
- Dropping single lines into a game engine or an app
- Handing individual takes to someone else
- Re-rendering one line without touching the rest
They are numbered in script order, so 0007.wav is always the seventh line, no
matter how many times you re-export.
The full mix is for review, not for finishing
The full mix is one continuous track of the whole render. It is the fastest way to listen back and catch a line that came out wrong, and it is what you send to a client who just wants to hear it. Do not build the final edit on top of it — you lose the ability to nudge a single line.
Subtitles and the timing sheet
The SRT is timed against the render, so it stays in sync as long as you do not move the clips. If you do move them, re-export rather than shifting the SRT by hand.
The CSV is the one people overlook. Every clip with its in and out point, as a spreadsheet. It is the fastest way to answer "how long is the narration for chapter four?" without opening an editor at all.
Audition markers
If you are doing the sound pass in Audition, the marker file drops straight into a session and gives you a labeled point for every line. Useful when you are mixing music underneath and need to know where the dialogue lands.
A note on target duration
If you set a target duration on a line — a segment longer than the audio itself — every export respects it. The WAV is padded, the SRT timing reflects the segment rather than the speech, and the timeline leaves the gap. That is the point: you are describing the shape of the edit, not just the sound.