Glossary

An accent is the particular way someone pronounces a language — the vowels, consonants, rhythm and stress that can mark out where a speaker is from, their background, or whether the language is their first. It differs from a dialect, which also takes in vocabulary and grammar; an accent is about pronunciation alone.

In media, accessibility means making audio-visual content usable by people with hearing or sight loss, mainly through closed captions, subtitles for the deaf and hard of hearing (SDH) and audio description. Broadcasters and many public bodies are legally required to meet accessibility standards.

Our subtitling services→

ADR (Automated Dialogue Replacement) is the process of re-recording an actor’s lines in the studio after filming, in sync with the picture. It is used to fix poor location sound, reflect script changes, or replace the original dialogue with a different performance or another language.

Our ADR services→

An AI or synthetic voice is speech generated by software rather than recorded by a person. Text-to-speech engines can produce audio quickly and cheaply, but they still lack the warmth, nuance and direction of a professional human voice actor, so brands tend to use them for drafts, internal notes or high-volume, low-stakes content.

Our AI voice over services→

Animation creates the illusion of movement by showing a fast sequence of images — hand-drawn, 2D or 3D — each slightly different from the last. It is used across film, television, games and explainer videos, and almost always relies on voice over and sound design to bring the characters to life.

Audio description is an extra narration track that describes the important visual details of a film, programme or live event — action, scene changes, on-screen text and body language — during the natural pauses in the dialogue, so blind and partially sighted audiences can follow what is happening.

An audio guide is a recorded commentary — usually played on a handheld device or an app — that leads visitors around a museum, attraction or outdoor route, adding context and background to what they are seeing. Audio guides are often produced in several languages.

Audiobook narration is the recording of a book read aloud by a single narrator or a full cast. It calls for consistent pacing, clear character differentiation and real stamina across many hours of audio, followed by careful editing to remove noise, breaths and stumbles.

Hire an audiobook narrator→

Audiovisual translation is the translation of content that accompanies a film, video or presentation — such as a voice over script, narration or subtitles. Unlike plain document translation, it has to work within timing and on-screen constraints, so the wording is adapted to fit the picture and the pace.

An audition is a sample performance a voice actor gives so a client can judge whether they are right for a role — either reading a prepared piece or a short section of the actual script.

A back translation is when an already translated text is translated back into the original language by a second linguist who has not seen the source. Comparing the two versions is a way of checking that the meaning carried through, and it is often required for medical and pharmaceutical content.

Broadcast quality describes audio or video that meets the technical standards required for television, radio and streaming — clean, correctly levelled, free of background noise and delivered in the right file format. Every GoLocalise recording is produced to broadcast quality.

Cartoon voices are the exaggerated, characterful voices used for animated characters. They are most common in children’s content but also feature in adverts, games and corporate animations.

Character voices are the distinct voices a voice actor creates to play different roles — from a warm corporate narrator to a villain in a game — adjusting tone, pitch and delivery to suit each part.

Characters per second is a measure of how fast a subtitle asks the viewer to read: the characters on screen divided by the seconds the subtitle is displayed. Most adult specifications sit between 15 and 20 CPS, and children’s content lower.

It is the single most common reason a subtitle file is rejected. Translations usually run longer than the original, so a faithful translation that is not condensed will breach the limit — condensing without losing the meaning is the skill being paid for.

Check your file’s reading speed→

CJK is shorthand for Chinese, Japanese and Korean, grouped together because they share challenges in software, typesetting and localisation, such as large character sets and complex text layout.

CJKV means Chinese, Japanese, Korean and Vietnamese — the main East Asian languages, grouped together for the extra handling they need in software and localisation.

Closed captions are on-screen text of the dialogue and relevant sounds that the viewer can switch on or off. Unlike plain subtitles, captions also flag who is speaking and note important sound effects, which makes them essential for accessibility and popular for sound-off viewing on social media.

Our subtitling services→

See Machine translation.

Computer-assisted translation (CAT) is human translation supported by software — tools that store previous translations, manage terminology and speed up repetitive work — while a professional linguist still makes every translation decision. It is not the same as machine translation, where the software does the translating.

We treat every script, document and audio file our clients send us as strictly confidential, and never share them with any third party.

In consecutive interpreting the speaker pauses every few sentences to let the interpreter relay what was said into the other language. It suits meetings, interviews and smaller events where a natural back-and-forth is fine and no special equipment is needed.

Copywriting is the craft of writing words that promote a product, service, brand or idea and prompt the reader to act. It runs through adverts, websites, scripts and marketing of every kind, and usually needs careful adaptation — not just literal translation — when taken into another language.

Corporate presentations are the videos and slide decks companies use to introduce themselves, explain their products or services, train staff or pitch to clients — often paired with a professional voice over.

Corporate training is the development of employees’ skills and knowledge through structured learning. Training content — from onboarding to compliance — is frequently produced as e-learning and video, and localised so international teams get the same material in their own language.

Corporate video production is the making of video content for a company’s own use — promotional films, training and safety videos, product explainers and results updates — usually aimed at a specific business or internal audience.

A clear voice over and clean, broadcast-quality sound are central to a polished corporate video, and international versions often add subtitles or a localised voice track.

Desktop publishing (DTP) is the use of page-layout software to design finished documents such as brochures, manuals and signage. In localisation, DTP is the step where a translated text is fitted back into the original design — allowing for text that expands or contracts, and for different scripts — so the foreign-language version looks as good as the source.

A dialect is a variety of a language particular to a region or social group, differing in vocabulary, grammar and pronunciation. Where the difference is only in pronunciation, the right term is accent rather than dialect.

See Literal translation.

A person’s dominant language is the one they are strongest in — usually their native language, and the one they should record or translate into for the most natural result.

Dub is short for dubbing: replacing the original dialogue with a new recording in another language, performed to picture so it fits the actors on screen. The word comes from doubling — the dialogue is doubled in a second language.

A dub costs several times what subtitles cost and takes longer, because it needs casting, direction, a recording session per language and a re-mix. What it buys is a viewer whose eyes are free, which matters for children’s content, action, and anything watched on a second screen.

Our dubbing services→

Dubbing is the post-production replacement of the voices on a film or video, most often to present it in a different language. It ranges from simple voice replacement to full lip-sync, where the new dialogue is matched to the actors’ mouth movements and timing.

Dubbing is standard for films, series and animation released in foreign markets, and calls for talent and directors who specialise in it.

Our dubbing services→

E-learning is training and education delivered electronically — online courses, modules and interactive lessons that combine text, images, audio, animation and video. It can be self-paced or instructor-led, and is often localised into many languages with voice over and translated on-screen text.

Our e-learning voice over→

E-learning translation, sometimes called courseware translation, is the adaptation of online training into other languages. Because e-learning is media-rich, it involves far more than translating text — it also covers voice over, on-screen graphics, DTP and syncing everything back into the course so it works just as well in each language.

EBU-STL is the subtitle format used by UK and European broadcast, defined by the European Broadcasting Union. It is a binary format carrying positioning and broadcast-specific fields that a plain SRT cannot hold.

Economic translation is the translation of economics and finance content, from academic papers to sales and banking reports. It relies on specialist vocabulary and subject knowledge, so it is handled by linguists who genuinely understand the field.

An explainer video is a short, engaging video — often animated — that quickly shows what a product, service or idea does and why it matters. A clear, well-cast voice over is usually what makes the message land.

A fansub is an unofficial subtitle track made by fans rather than by the rights holder. It is not a licensed deliverable and is not cleared for distribution.

Worth naming because fansub conventions — translator’s notes on screen, honorifics left untranslated, karaoke-styled opening credits — are sometimes asked for on licensed work, and most platform specifications forbid them.

FIGS stands for French, Italian, German and Spanish — usually the first four languages a company localises into when entering the European market.

Financial translation is the translation of banking, investment and insurance material, where accuracy is critical. It is carried out by specialist native-speaker translators who work only into their mother tongue and know the financial field.

Foley is the studio recreation of everyday sound effects — footsteps, rustling clothes, clinking glasses — performed in time with the picture so a soundtrack feels real and immersive.

Forced narratives are subtitles that appear even when the viewer has subtitles switched off — an on-screen sign, a written note, or a line spoken in a language different from the rest of the title.

They are delivered as their own track. Platforms check for them, and a title that needs one and ships without it fails quality control.

Frame rate is how many frames of picture there are per second, and it is the grid a subtitle file’s timings are locked to. The same file played at 23.976 and at 25 frames per second drifts progressively out of sync.

Confirm the frame rate of the exact master the subtitles will play against, not of the edit they were made from.

A freelance translator is a self-employed translator who works independently rather than in-house — sometimes directly for clients and sometimes through an agency.

A freelancer is a self-employed professional who takes on work for different clients rather than committing to a single employer. Translation and voice over are both fields where freelancing is common.

A gist translation is a rough translation — done by a person or by machine — that conveys the general meaning of a text without the polish of a full translation. It is useful when you only need to grasp the essence of something quickly.

Globalisation is the growing integration of economies, cultures and communication across the world through trade, technology and the movement of people and ideas. For businesses it usually means adapting products and content for many markets — which is where localisation comes in.

Hardcoded or burned-in subtitles are permanently rendered into the video image, so they always display and cannot be switched off. They are useful when you cannot rely on a player supporting separate subtitle files, but a new version is needed for each language.

Our subtitling services→

Human translation is carried out by a professional linguist rather than by software. It captures meaning, tone, idiom and cultural nuance that machine translation still misses, and is the right choice for anything customer-facing or high-stakes.

An idiom is a phrase whose meaning cannot be worked out from the individual words — like ‘break the ice’ or ‘under the weather’. Idioms rarely translate literally, which is why they need careful adaptation rather than word-for-word translation.

In-store announcements are the recorded or live messages played over a shop or shopping-centre PA system — from promotions to information and safety notices for customers.

An infomercial is a long-format television advert — typically around half an hour — that presents a product in detail, blending information with selling. The term is sometimes stretched to cover any long, persuasive video presentation.

Internationalisation is designing a product, website or piece of content from the outset so it can be adapted to different languages and regions with minimal rework — for example allowing for text expansion and keeping text separate from code. It is the groundwork that makes later localisation faster and cheaper.

Interpreting is the spoken or signed transfer of meaning from one language to another in real time, as opposed to translation, which deals with written text. The two main modes are simultaneous and consecutive interpreting.

ISDN is an older digital phone standard once widely used to record voice talent remotely in real time between studios. It is being phased out in favour of internet-based systems such as Source-Connect.

iTT is Apple’s profile of TTML, required for delivery to the Apple platforms.

IVR (Interactive Voice Response) is the automated telephone system that greets callers and guides them through spoken menus — ‘press 1 for sales’ — using recorded voice prompts. A clear, professional IVR voice shapes a caller’s first impression of a business.

A jingle is a short, catchy piece of music written for an advert, usually carrying a slogan or brand name. It is a form of sound branding designed to lodge in the listener’s memory.

Kerning is the fine adjustment of the space between individual letters so text looks even and legible. It matters in multilingual desktop publishing, where accented characters and different scripts can throw off the spacing of a translated layout.

Legal translation is the translation of contracts, court documents, statements and other legal texts. Because law is tied to each country’s own system and a mistranslation can have serious consequences, it is handled only by specialist legal translators.

Lip sync is the craft of matching dubbed dialogue to the mouth movements and timing of the on-screen actor, so the new language looks as though it was spoken by the original performer. It is the most demanding form of dubbing.

Our dubbing services→

Lip-sync dubbing is dubbing in which the new dialogue is written and performed to match the mouth movements and timing of the actor on screen, so that the replacement reads as the original performance.

It is the most expensive form of dubbing, because the translation has to be rewritten to fit the mouth as well as the meaning, and every line is recorded to picture.

Our dubbing services→

A literal translation renders a text word-for-word rather than conveying its overall sense. It can be useful for reference, but it often reads awkwardly or misses the meaning, which is why most content calls for a more natural approach.

Literary translation is the re-creation of novels, poetry, plays and other literary works in another language. Beyond the meaning, it has to carry the style, rhythm and artistry of the original, which makes it one of the most demanding forms of translation.

A loanword is a word taken from one language and adopted into another — like ‘café’ or ‘kindergarten’ in English.

A locale is a specific combination of language and region — such as French (France) versus French (Canada) — together with its conventions for dates, numbers, currency and spelling. Content is usually localised per locale, not just per language.

Localisation is adapting a product, website or piece of content for a specific market — not just translating the words, but adjusting tone, images, formats, currency and cultural references so it feels made for that audience.

The same language can need different localisation for different regions: Spanish for Spain reads differently from Spanish for Latin America, and word choices in English vary from one country to the next.

An M&E track is the mix of a title with the dialogue removed, leaving music and sound effects. It is what a dub is laid onto.

If the rights holder cannot supply one, the soundtrack has to be rebuilt before dubbing can start, which is a separate job with its own cost and schedule. Ask for it early.

Machine translation (MT) is translation produced automatically by software. It has improved a great deal and is useful for speed and volume, but it still struggles with nuance, idiom and tone, so important content is either translated by a person or post-edited by one.

MTPE is the process of having a professional linguist review and correct machine-translated text to bring it up to human quality. It can speed up large, lower-risk projects, though the amount of editing needed depends on the content and the quality of the raw output.

Market research translation is the translation of surveys, questionnaires and findings for international research. Accurate wording and local idiom are vital so responses stay comparable across markets, so it is done by specialist native-speaker linguists who understand research.

Marketing translation is much more than literal translation. Slogans, wordplay and campaign copy have to be adapted so they carry the same appeal and emotion in the new language and culture — closely related to transcreation.

Medical translation is a highly specialised field covering clinical, pharmaceutical and healthcare content. It is handled only by qualified translators with medical expertise, since errors can have serious consequences.

A person’s mother tongue is their first or native language — the one learned from birth and usually spoken most naturally. A second language is any language learned in addition to it.

Multimedia is content that combines several formats — text, audio, images, animation, video and interactivity — rather than a single medium. Most e-learning, corporate and web content today is multimedia, which is why localising it involves voice, text and design together.

A narrator is the voice that tells a story or guides the audience through content — in a film, documentary, audiobook, e-learning course or corporate video. The narrator is often a specialist voice actor rather than a character seen on screen.

A native speaker learned a language from birth and uses it with complete fluency and cultural intuition. GoLocalise casts native-speaker voice actors and translators only, so the read and the wording sound natural to the target audience.

A neutral accent is a delivery that avoids strong regional markers so it sounds natural to the widest possible audience within a language — for example a neutral Latin American Spanish, or a neutral General American English for international projects.

On-hold messaging is the recorded voice and music callers hear while waiting on the phone or being transferred. Businesses use it to keep callers informed and reinforce their brand rather than leaving dead air.

Open captions are text rendered permanently into the picture rather than carried as a separate file. They cannot be switched off, restyled or swapped for another language.

They are what social platforms and in-feed video usually need, because the player has no subtitle track. The trade-off is that one burnt-in language means one video file per language.

A passive speaker understands a language well — usually from childhood exposure — but has limited ability to speak it. Because their comprehension is already strong, passive speakers can often regain active fluency quite quickly.

Pick-ups are small sections re-recorded after the main session to fix a mistake, reflect a script change or add a line, matched to the original recording in tone, level and studio setup so the join is seamless.

Post-production covers everything that happens after shooting or recording — editing, voice over, dubbing, mixing, sound effects and final delivery. It is the stage where the raw material is turned into a finished piece.

A PowerPoint presentation is a set of slides built in Microsoft PowerPoint to present information on a topic. Presentations are often narrated with a voice over or translated for international audiences.

A pronunciation guide tells the voice actor how to say tricky names, brands, acronyms and foreign words, so the read is correct and consistent. It is especially important for product names and technical or medical terms.

Proofreading is the final check of a text for errors in spelling, grammar, punctuation and formatting before it is published or delivered — the last quality step in translation and DTP.

Quality assurance is the set of final checks that make sure a delivery is right before it reaches the client. In localisation, linguistic quality assurance (LQA) is a review by a qualified linguist to catch errors in language, terminology, formatting and cultural fit.

A radio promo is a short recorded announcement used to advertise or promote something on the radio, produced by professional voice talent.

Received Pronunciation — sometimes called ‘BBC English’ — is the traditional standard accent of British English. Widely understood and neutral in register, it is a common choice for corporate, e-learning and documentary voice overs aimed at UK or international audiences.

British voice over→

A recording studio is a purpose-built, acoustically treated space for capturing and mixing clean, broadcast-quality audio. It typically has a live room where the talent performs and a control room housing the recording equipment, sometimes with isolation booths for louder sources. GoLocalise records in professional studios around the world.

Respeaking is a way of producing live subtitles: a trained operator listens to a broadcast and re-speaks it, including punctuation and speaker changes, into speech-recognition software that turns it into captions on screen in near real time.

A retake is a re-recording of part or all of a voice over, carried out at the client’s request if they are not happy with the original delivery.

A rythmo band is a technique used in French and some other dubbing traditions in which the adapted dialogue scrolls under the picture in sync with the mouth, so the actor reads and watches at the same time.

SCC is the closed-caption format of US broadcast, encoding captions as the raw data a caption decoder reads. It is frame-accurate and carries positioning.

Script translation covers far more than films — documentaries, interviews, adverts, stage plays, corporate videos and more. A good script translation carries the wit, tone and subtleties of the original across the cultural gap, not just the literal words.

SDH are subtitles that include not only the dialogue but also speaker identification and important non-speech sounds such as music or a door slamming. They combine the accessibility information of closed captions with the on-screen styling of subtitles.

Our subtitling services→

Shot changes are the cuts in the picture. A subtitle that runs across a cut makes the viewer start reading it again, so most specifications require subtitles to begin and end with the shot rather than run through it.

A simuldub is a dub released alongside the original broadcast, or within days of it, rather than a season later. It means recording each episode as it airs instead of after the run finishes.

The schedule is the whole difficulty: scripts arrive late, the cast has to be held available every week, and there is no time to fix an episode once it has shipped.

In simultaneous interpreting the interpreter renders the speaker’s words into another language in real time, usually from a soundproof booth with headsets for the audience. It is the format used at conferences and large multilingual events.

Sound effects are recorded or created sounds — footsteps, explosions, ambient noise — added to film, video, games and other media to make a scene feel real or to underline a moment. They are handled as a separate layer from dialogue and music.

Mixing is the stage where dialogue, voice over, music and sound effects are balanced, levelled and blended into a single polished soundtrack that sits correctly across different playback systems.

The source language is the original language of a text or recording — the one being translated from.

Source-Connect is a system for recording voice talent remotely in real time over the internet at broadcast quality, letting a client or director in one studio direct an actor in another anywhere in the world.

Spotting is deciding where each subtitle starts and ends against the audio and the picture. It happens before translation, and it is what separates a subtitle file from a transcript.

An SRT is one of the most common subtitle file formats: a simple text file that pairs each subtitle with its start and end timecodes. Almost every video platform and player accepts SRT files.

Our subtitling services→

A style guide is a reference document that sets out how a brand’s content should read in a given language — tone of voice, preferred terminology, spelling, formatting and things to avoid — so that every translator and voice actor stays consistent.

Sub is short for subtitled: a title released or watched in its original audio with translated subtitles on screen. In anime it means the original Japanese voice track with subtitles in the viewer’s language.

Subtitling costs less than dubbing and is faster, and it keeps the original performances intact — which is why it is the default for festival releases, limited runs and anything where the original cast is part of the draw.

Our subtitling services→

Subtitling is the process of adding timed on-screen text, usually a translation of the spoken dialogue, to video. Good subtitling condenses speech to a comfortable reading length and speed while keeping the meaning and tone intact.

Our subtitling services→

The target language is the language a text or recording is being translated into.

Technical translation is the translation of specialised material such as manuals, user guides and engineering or scientific documents. It needs strong subject knowledge and consistent terminology, so it often uses translation memories and termbases and is handled by expert linguists.

A television commercial is a paid advert broadcast on television, ranging from a few seconds to a couple of minutes. A strong voice over is often what carries the message and the brand.

Terminology management is the practice of agreeing a client’s key terms and their approved translations and storing them in a database, called a termbase, so the same words are used consistently across every project and language.

Text-to-speech is software that converts written text into spoken audio using a synthetic voice. It is helpful for accessibility and high-volume content, but does not match a professional human voice for brand, emotion and nuance.

Timecode is the running hours:minutes:seconds:frames reference stamped on video and audio. It lets everyone on a project point to the exact same moment, which is essential for subtitling, dubbing and syncing a voice over to picture.

Timed text is the umbrella term for text that carries its own timings — subtitles, captions and SDH alike. Platform and broadcaster requirements are usually published as timed-text specifications.

A trailer is a short promotional preview of an upcoming film, series or game, edited to build anticipation. Trailer voice overs have a distinctive, high-impact style of their own.

Transcreation is the creative adaptation of a message from one language to another so it keeps the same intent, tone and emotional impact rather than the same words. It is used for advertising, slogans and marketing, where a literal translation would fall flat.

Our translation services→

Transcription is the process of turning spoken audio into written text. It should not be confused with translation, which changes the language, or with transliteration, which changes the writing system.

Our transcription services→

A translation memory is a database that stores previously translated sentences so they can be reused when the same or similar text comes up again. It improves consistency and reduces the cost and turnaround of updates and repetitive content.

Transliteration is the mapping of text from one writing system into another, letter by letter — for example rendering Cyrillic or Arabic script in the Latin alphabet — so it can be read by people unfamiliar with the original script. It deals with the spelling and characters, whereas transcription deals with the sounds.

TTML is an XML-based timed-text standard, with IMSC as the profile the streaming platforms generally require. It carries full styling, positioning and language metadata.

Turnaround time is how long a project takes from order to delivery. Most short voice over jobs are delivered within 24–48 hours; larger or multilingual projects are scheduled with your project manager.

A tutorial is a step-by-step guide that teaches how to do something, usually by example. Tutorials range from simple written instructions to interactive, narrated video walkthroughs.

Typesetting is the arrangement of text and type for display or print — choosing and positioning the characters so the result is even and readable. In localisation it ensures translated text sits correctly, including languages with accents or entirely different scripts.

A UN-style voice over, also called a voice replacement, plays the translated narration over the original audio, which stays faintly audible underneath — the style familiar from documentaries and news interviews. It is faster and cheaper than full lip-sync dubbing.

A video game is an interactive electronic game played on a platform such as a console, computer or phone. Games often require voice over, localisation and audio for many characters and languages.

Video production is the whole process of planning, filming, editing and delivering a finished video — from a single camera operator to a full crew. It spans commercials, corporate films, events and more, with voice over and sound as key ingredients.

A virtual tour is a simulated walkthrough of a real place, built from a sequence of images or video and often enhanced with narration, music and sound effects. Multilingual narration makes tours accessible to international visitors.

A voice actor is a performer who uses only their voice to bring a script to life — narrating, characterising or selling — across commercials, corporate video, e-learning, animation, games, dubbing and more.

Browse our voice actors→

A voice demo is a short showreel a voice actor uses to showcase their range, usually featuring a mix of styles such as corporate, commercial, e-learning and narration. If you are sending a demo to GoLocalise, include a good variety of samples — and, if you record in English with a foreign accent, a sample of that too.

Voice matching is casting a new voice actor whose tone and delivery closely resemble a specific original voice — used, for example, to continue a campaign, re-record content or dub a well-known character into another language.

A voice over is a recorded voice that is heard but not seen on screen — narrating, explaining or selling over a video, advert, e-learning course, game or presentation. It may be delivered by a specialist voice actor or by someone featured elsewhere in the production.

Browse our voice actors→

VTT is the subtitle format of the web and of live streaming, supported natively by browsers. It extends SRT with basic styling and positioning.

Website localisation is adapting a website for a new market’s language and culture — combining translation with technical and design changes so everything, from wording to images to units and formats, feels natural to local visitors.

It goes well beyond translating the text alone: the aim is a site that reads as though it was created for that market in the first place.

Word count is the number of words in a script or document. It is the main factor when quoting translation and, together with running time, when estimating voice over recording and pricing.

XLIFF (XML Localization Interchange File Format) is a standard file format for passing translatable text between tools and translators. It keeps the content separate from the surrounding code or layout, so translations can be dropped back into a product cleanly.

Adding accurate subtitles or captions to YouTube videos widens your audience, supports viewers watching with the sound off, and helps search engines understand the content. Uploading a professionally produced subtitle file is far more reliable than YouTube’s automatic captions.

Our subtitling services→

Zero translation is deliberately leaving a term in its original language rather than translating it — common with brand names, product names and established acronyms, where an equivalent would be unfamiliar or dilute the brand.