Translation, multilingual publishing and AI training data

Agencies, publishers and e-learning teams send us files to translate and typeset. AI companies send us data to label and test. Native speakers do the work, and a second person checks it before it goes back to you.

Offices in London, Florida and Gurugram

English sourceLive preview

Your course goes live in twelve languages on Monday.

Into

تنطلق دورتك التدريبية باثنتي عشرة لغة يوم الاثنين.

Arabic, set right to left

あなたのコースは月曜日に12の言語で公開されます。

Japanese, with Japanese line-break rules

आपका कोर्स सोमवार को बारह भाषाओं में लाइव होगा।

Hindi, Devanagari script

Tu curso se publica en doce idiomas el lunes.

Spanish, Latin American

你的课程将于周一以十二种语言上线。

Chinese, simplified

Ihr Kurs geht am Montag in zwölf Sprachen online.

German, about a third longer than English

귀하의 강좌가 월요일에 12개 언어로 공개됩니다.

Korean, Hangul

Seu curso entra no ar em doze idiomas na segunda-feira.

Portuguese, Brazil
Translated Laid out Checked by a second linguist

Giga Translations in short

Giga Translations is a London language services company. We translate documents, set them in print and digital layouts, transcribe and subtitle media, localize e-learning, and build multilingual data for AI teams.

A red London bus passing a pub on a London street
10+ years
in business
To confirm
200+
large interpreting projects completed
To confirm
3 offices
London, Florida and Gurugram, with team members across Europe, Asia and America
3 hours
typical email reply time on business days
To confirm

Client logos appear here once each client has approved their use.

Who we work with

Most of our work comes from four kinds of teams. Each one needs something different from us.

Language and publishing services

Five groups of services, all run by the same project team. Pick a group to see what's in it. AI training data has its own section.

Translation and Localization

Documents, websites and marketing copy in the languages your readers use, checked by a second linguist before delivery.

AI training data from native speakers

Models trained mostly on English data often miss how people actually write and speak in other languages. We collect, label and evaluate data with people who use those languages every day, and every project starts with a pilot batch on your own data.

An audio waveform on a dark screen, the kind of recording that becomes speech training data

How the quality stays measurable

  1. Pilot batch firstWe agree guidelines and gold examples with you, label a small batch and review it together before anything scales.
  2. Agreement checked on every batchAnnotators are compared against each other and against the gold set, and you get the scores with each delivery.
  3. Disagreements go to a senior reviewerUnclear items go to a senior reviewer for a decision, and the guidelines are updated so the same question doesn't come back.
EnglishNamed entities

Giga ORG sent the files to Madrid LOC on Friday DATE.

SpanishNamed entities

Giga ORG envió los archivos a Madrid LOC el viernes DATE.

A server rack with green status lights

Your data stays yours

  • Every annotator works under a signed NDA.
  • Access is limited to the people on your project.
  • We work inside your platform or deliver files in your format.
  • Your data is deleted when the project ends, unless you ask us to keep it.
Security setup and certifications to confirm

How a project runs

The same four steps whether you send one brochure or a hundred hours of audio.

  1. Send your files

    Upload files or share access, and tell us the languages, format and date you need.

  2. Get a quote and a date

    We check the files and send a fixed quote and delivery date. AI projects start with a pilot batch.

  3. We do the work and check it

    A specialist does the work. A second person checks it against the source before it leaves us.

    Second check QA
  4. You get finished files

    Delivered in the format you asked for, with notes on anything we'd like you to look at.

The word for language, in ten scripts

Right-to-left text, Asian scripts and Indic scripts each need their own fonts, line breaks and spacing. We set them by hand, and a native speaker checks the finished pages.

The same care goes into signs, packaging and screens, where a wrong line break is seen by everyone who walks past.

A Hong Kong street lined with shop signs in Chinese characters
  • LanguageEnglish, Latin script
  • لغةArabic
  • भाषाHindi, Devanagari
  • 言語Japanese
  • ЯзыкRussian, Cyrillic
  • שפהHebrew
  • 语言Chinese, simplified
  • ภาษาThai
  • 언어Korean, Hangul
  • ΓλώσσαGreek
Layout
InDesign, QuarkXPress, Illustrator, CorelDRAW, Publisher
Office
Word, PowerPoint, Excel
E-learning
Articulate Storyline, MadCap
eBooks
EPUB, Kindle, print-ready PDF
AI training data
Your labeling platform, or files in JSON and CSV To confirm

Apps, games and AR for learning content

We also build the products your content lives in: reading apps, learning games and AR that works from printed pages.

See Digital Development
  • eBook and learning appsReading and course apps for iOS and Android, built around your books and lessons.
  • Educational gamesGames, simulations and interactive assessments that teach and test a skill.
  • AR and VRAR markers on printed pages, virtual labs and 360-degree content.

Tell us about your next project

Book a call to talk it through, send us a message, or send your files and we'll reply with a quote.

Phone hours
Weekdays 9:30 to 16:00, weekends 9:30 to 12:00, UK time

Send us a message

We use your details only to reply to this message. See our Privacy Policy. To send files, use the quote request form.

Book a call

30 minutes with our project team. Choose a weekday and a time between 10:00 and 18:00 UK time.

Calls run Monday to Friday. Weekends aren't available.

Choose a date to see times