How to Prepare Your Script for Multilingual Voice-Over: A Practical Guide
Key Takeaways
- The quality of the source script directly determines the quality of the multilingual voice-over. A poorly written script produces poor translations and difficult recordings in every language.
- Scripts written for reading are different from scripts written for speaking — voice-over requires the latter.
- Language expansion is unavoidable: most European languages run 15–25% longer than English when spoken. Scripts that do not account for this create timing problems in every target language.
- Providing clear pronunciation guidance for names, brands, and technical terms prevents costly pick-up sessions after recording.
- A finalised, locked script before translation begins is the single most important process step. Changes after translation starts multiply costs across every language.
A multilingual voice-over project lives or dies by the quality of the source script. Translation and recording are downstream processes — if the script they are working from is poorly structured, written for the page rather than the microphone, or not finalised before translation begins, problems compound across every language version.
Table of Contents
ToggleThis guide sets out what makes a voice-over script work well across multiple languages, the preparation steps that save time and cost, and the common mistakes that cause delays and re-recording sessions.
Write for the Ear, Not the Eye
The most common source of problems in voice-over projects is a script written as a document rather than as spoken audio. Corporate copywriters, training developers, and marketing teams naturally write in a register suited to reading. Voice-over requires different writing.
What this means in practice:
- Use shorter sentences. Long, complex sentence structures that work on the page become difficult to deliver and follow when spoken. Aim for an average of 15–20 words per sentence.
- Avoid parenthetical clauses. A sentence like “The system — which was installed in 2022 and has since been updated — monitors all incoming data” works in print but is awkward to record and harder to follow when heard.
- Avoid abbreviations and acronyms without guidance. If an acronym should be read as letters (e.g. “GDPR” read as “G-D-P-R”), indicate this. If it should be read as a word (e.g. “NATO”), indicate that too. Different languages may have different conventions for the same acronym.
- Use active voice. Passive constructions are grammatically heavier and more tiring to listen to over a long recording.
- Read the script aloud before submitting it. If you stumble on a sentence, a voice artist will too — and so will translators adapting it into other languages.
Account for Language Expansion
English is a compact language. Most European languages expand when translated from English — sometimes significantly. Typical expansion rates:
| Language | Typical expansion vs English |
|---|---|
| French | +15–20% |
| German | +20–30% |
| Spanish | +15–25% |
| Italian | +15–20% |
| Portuguese | +15–25% |
| Polish | +20–30% |
| Russian | +10–15% |
| Japanese | -10–20% (contracts) |
| Chinese (Mandarin) | -20–30% (contracts) |
In a video with fixed timing — where the voice-over must fit a specific segment duration — this expansion creates a problem. A 30-second English narration segment translated into German and recorded at the same pace will run 36–39 seconds, overrunning the video by 20–30%.
How to handle this:
- Build timing tolerance into the video during production. If a segment has a 30-second narration slot, give the voice-over 25 seconds of actual content and leave 5 seconds of buffer.
- Instruct translators to adapt for time, not just translate. A good voice-over translator does not translate word for word — they produce a script of equivalent meaning that fits the timing. Brief them explicitly on any timing constraints.
- For Asian languages that contract rather than expand, the opposite problem applies — narration may end before the visual sequence does. Brief translators on minimum as well as maximum duration.
Lock the Script Before Translation Begins
This is the most important process rule in multilingual voice-over, and the one most commonly broken.
If the source script changes after translation has begun — even small changes — those changes must be cascaded across every language being translated. A one-line amendment to the English script may require amendments to ten translated scripts, followed by ten re-recorded segments, followed by ten re-edited video files.
The cost and time impact of mid-project script changes in a ten-language project can be ten times greater than the same change in a single-language project. In practice, late changes are one of the most common causes of budget overruns and missed deadlines in multilingual video production.
Before translation begins, confirm:
- The script has been reviewed and approved by all internal stakeholders
- Any legal or compliance review has been completed
- Product names, company names, and key terminology have been finalised
- The video edit that the voice-over will be recorded against is locked
Provide a Pronunciation Guide
Voice-over translators adapt the script and voice artists record it — but neither can know how your organisation pronounces its own product names, brand names, executive names, or specialist terminology unless you tell them.
A pronunciation guide is a simple document listing every name, brand, acronym, and technical term in the script, with phonetic guidance on how it should be pronounced. It should specify:
- Product and brand names (especially if they are unusual or not in the target language)
- The names of any individuals mentioned or quoted
- Technical terms that may have alternative pronunciations
- Acronyms — whether to read them as letters or as words
- Any foreign-origin terms that may be mispronounced in the target language
Providing this guide before recording begins prevents the most common source of pick-up sessions — recordings that must be redone because a name or term was pronounced incorrectly.
Format the Script for Recording
A script formatted for recording makes the voice artist’s job easier and reduces the likelihood of errors, retakes, and inconsistencies across a long recording session.
Script formatting best practice:
- Use a two-column format: timecode or scene reference in the left column, script text in the right. This allows the voice artist and director to navigate the script efficiently.
- Mark pauses explicitly where the visual requires a beat — do not leave the artist to guess.
- Use bold for emphasis where a word should carry stress. This is especially important in translated scripts where the natural stress pattern of the target language may differ from English.
- Mark any specific delivery notes — “upbeat”, “slower here”, “authoritative tone” — clearly in the script rather than explaining them verbally during the session.
- Number every line or segment. This allows pick-ups to be referenced precisely (“re-record line 47”) rather than described (“the bit about the installation process”).
Brief Your Translation Agency Fully
A voice-over script translation is different from a document translation. The translator needs to understand:
- The format of the final output (narrator voice-over, lip-sync dubbing, subtitles used alongside)
- Any timing constraints per segment
- The tone and register required (formal, conversational, authoritative, warm)
- The target audience in each language
- Any terms that must remain in English (brand names, product names, technical terms that are not translated)
- Whether the translation should be a close adaptation of the English or a freer localisation
Providing this brief — ideally in a written document — before translation begins produces a better result and reduces the need for revision cycles.
Review the Translated Script Before Recording
For high-value or high-stakes content, build in a native-speaker review of the translated script before recording begins. This review should check:
- That the translation reads naturally when spoken aloud
- That timing constraints have been respected
- That the tone and register match the brief
- That all names and terms have been handled correctly
A script review before recording is significantly less expensive than a re-recording session after delivery.
Summary
A well-prepared source script — written for the ear, timed with expansion in mind, locked before translation begins, and accompanied by a pronunciation guide — dramatically reduces the cost, time, and revision cycles in a multilingual voice-over project. The preparation investment at the script stage pays back many times over in smoother production across every target language.
Global LTS manages multilingual voice-over projects from script translation through to final delivery, including voice casting, studio recording, and post-production. We advise clients on script preparation as part of every project briefing. Contact us to discuss your voice-over requirements.


