Textless Masters and Elements: Exporting a Clean Picture for Localization

What textless means, and what it does not

Textless refers to text added during editing and graphics work. A textless version removes:

• Title cards and chapter titles.
• Lower thirds with names, roles and locations.
• Burned-in captions and subtitles.
• Callouts, labels, arrows with text and annotations.
• Text inside motion graphics, charts and end cards.

It does not remove text that was in front of the camera: street signs, product packaging, a whiteboard, a phone screen, an app interface in a screen recording. That text is part of the footage and only changes by reshooting, re-recording or visual effects work. How to handle it in translated versions, from subtitles to overlays, is covered in translating on-screen text.

Textless masters versus textless elements

In professional delivery, two forms are common:

Textless master

A full-length copy of the finished video with all post-production text off. Simple to use: drop it on a timeline and add translated text.

Textless elements

Only the shots that have text over them, in clean form, often with a second or two of extra picture either side. Commonly delivered after the end of the main programme on the same file, separated by a short gap.

For creators, a full textless master is usually the more practical choice. Storage is cheap compared with editing time, and a full-length file can be lined up with the original in one step. Textless elements make sense for long productions where only a few shots carry text.

Why a clean picture makes localization easier

Without a textless version, every translated graphic has to sit on top of the original. That leaves a few unattractive options: cover the English title with a box, blur it, or leave it and add a translated subtitle beneath it. All of them look like a patch.

With a textless master, translated titles and labels go onto the picture exactly as the originals did, in the same position and style. Other benefits follow:

• Burned-in captions stop being a problem. A video with English captions baked in cannot carry another language's subtitles without two sets of text on screen.
• Translated text can be resized. Text in many languages runs longer than English, and German or Finnish titles may not fit the original layout; see text expansion in translation. On a clean picture you can reflow it.
• Right-to-left languages can be laid out properly, mirrored where the design calls for it.
• Corrections are simple. A typo in a lower third can be fixed without rebuilding the whole video.
• Re-cuts reuse the same base, whether for another platform, a shorter edit or an updated version.

What to export alongside the textless master

A clean picture is only half the package. Whoever builds the translated version also needs to know what the text was and how it looked:

• A text list: every on-screen text in order, with its start and end time and its position, in a spreadsheet or document. Translators work from this.
• The graphics source files: motion graphics templates or project files for titles and lower thirds, so they can be edited rather than recreated.
• Fonts, or the names and licenses of the fonts used. Check the font license covers the people you share it with, and that the font supports the target language's characters.
• Brand colours and any style notes for the graphics.
• The music-and-effects stem and a dialogue stem, so the soundtrack can be rebuilt too; see the M&E track.

The text list is easy to generate at the end of an edit and painful to reconstruct from a finished video. Make it part of the export routine.

How to export a textless version

The export is trivial if the timeline is organized for it:

1. Keep every text element on its own top tracks in the timeline, above all footage and effects.
2. Build titles and lower thirds as separate layers or templates rather than baking them into graphics exported as finished video.
3. For graphics that combine animation and text, ask the designer for a version with the text layer off.
4. When the edit is locked, disable the text tracks and export at the same resolution, frame rate, length and codec as the master.
5. Compare the two exports side by side at a few points to confirm every text element is gone and nothing else changed.
6. Name it to match the master with a textless tag, for example 2026-10-02_lens-guide_textless_v2.mov.

If you burn captions in for social platforms, always export a version without them first. Burned-in captions are the most common reason a video cannot be cleanly localized later.

Example: a product explainer in three languages

Managing textless files over time

A textless master is only useful if it matches the current version of the video:

• Version them together. Whenever the master changes, export a new textless version with the same version number, even if only the text changed.
• Store them together. Keep the textless master, text list and graphics files in the same folder as the master, not in a separate archive that drifts.
• Archive the graphics project. A textless master with no editable titles still means rebuilding every graphic.
• Record which language versions were built from which textless version, so you know what to rebuild when the original changes.

A broader approach to folders and names is described in making videos translation-ready, which also covers planning text out of the picture from the start.

Limits and trade-offs

• Storage: a textless master roughly doubles the space a finished video takes in your archive.
• Not everything can be textless. Text in the scene, interface text in screen recordings and text designed into animated backgrounds stay unless the source is rebuilt.
• Old videos are hard to fix. If only a flattened export with burned-in text exists, removing it means masking, cropping or object-removal tools, which can leave visible artifacts.
• More versions to manage. Each language now has its own timeline and graphics set; without a naming and versioning rule, mistakes creep in.
• Extra editing time per language. A clean picture makes translated graphics possible, not automatic; someone still has to lay them out.

mydubly and the picture

mydubly never alters the picture. It does not remove, detect or translate on-screen text, and it does not burn subtitles into video. When it produces a translated video, it copies the original picture packet by packet without re-encoding and replaces the audio with the dubbed mix: the translated voice over the original background, with the original speech removed. That means what you upload is what you get back visually: upload a texted master and the English titles stay; upload a textless master and the result is clean picture with the translated voice.

For most localization work, the useful outputs are the translated audio, which goes onto your textless timeline in your editor (as it is if you dubbed the full mix, or with the M&E if you dubbed the dialogue stem), and the SRT or VTT subtitle files, which you add as separate tracks or style and burn in yourself. Translated text for titles and lower thirds comes from your text list and a translator or reviewer, not from mydubly. The audio side is handled in AI dubbing; the full process of adapting a video for another market is described on the video localization page.

Next step

On your next project, put all text on top tracks, export a textless master and a text list with the final master, and file them together. For campaign videos where graphics carry the message, the marketing video translation page shows how the pieces fit together.