SCRIBECUE
untitled.srt / no media
Drop a video or audio file here
or press ⇧O — files never leave your machine
00:00:00.000 / 00:00:00.000

What should happen to this script's timings?

You can undo either change with Z, and change this in the text to speech panel.

Welcome to Scribecue

Scribecue is a keyboard-first editor for audio description scripts. Your video and your script stay on your computer: nothing is uploaded.

This quick setup picks the tools you need. You can change any of it later, and run it again with Setup at the top.

Sounds like

Tell text to speech how to say words in this block. Only the reading changes: the block's text stays as written, in the list and in exports.

Spell it how it sounds, such as shiv-AWN for Siobhan. Each word or phrase is matched as a whole word, in any case. Tip: select a word in the text box before pressing ⌥⇧P to add it.

Text to speech

Hear blocks read aloud, to check where each line sits and whether it fits before the next piece of dialogue.

Applies to blocks you create with ↵ or N. Blocks marked with H, drawn on the timeline, opened from a file, or whose end you've set yourself keep their timing.

Drag an edge, or use A, S, Q, W, E or R, and the reading speeds up or slows down to fill the block, shown as a custom speed. Picking a speed for the block yourself stops this.

Update all timings

Set the end of every block from how long its reading takes, with each block's own speed. This includes blocks marked with H and blocks opened from a file. Ends never run into the next block, and Z undoes it.

Fit speeds to timings

The other way round: every block keeps its timing, and its reading speeds up or slows down to fill it, between −50% and +100%. Useful for a script opened from a file. Blocks with a speed you picked keep it, and Z undoes it.

Export description track

Save every block's reading as one WAV file, each starting at its block's start time, ready to mix with the programme. Readings peak at −8 dBFS, leaving headroom for processing and mastering.

Online voices

Online voices often sound more natural, but each block's text is sent to the company that runs the voice, such as Google or Microsoft, to be spoken. Leave this off if your script mustn't leave your computer. The voices built into your computer keep everything on it.

Keys

V read the selected block from its start · ⇧V set the block's end to where the reading finishes · [ ] this block slower / faster · ⌥⇧P how words in this block sound · ⌥⇧V this panel

Check for updates?

Scribecue can check GitHub once a day and tell you when a new version is out.

GitHub sees the request and your IP address, like any website visit. Nothing about your files or text blocks is sent. If you say no, Scribecue never contacts anything outside your computer.

You can change this later by clicking the version number in the bottom right.