Captions, subtitles or transcripts: which do you need?

Choose captions when viewers need to follow the speech in the same language. Choose translated subtitles when part of your audience speaks another language. Choose a transcript when people will want to read, search or quote the words later. Many events need more than one.

What is each output?

People often use “captions” and “subtitles” for the same thing, and that is fine. For a quote, what matters is the job: same language or another language, and which language.

How do you decide which ones you need?

Start with the people who will watch, not the file types. Ask:

1. Who cannot follow the video as it is? Guests with hearing loss, older relatives, staff who work in another language, or viewers on a phone in a public place.
2. In which languages is the event spoken? One language, or a mix, such as Malay and English in the same sentence.
3. What will people do with the words? Watch along, search for a name, or copy a quote into a report.

For example, a company town hall in English and Malay for staff across the country may need captions in both languages, plus a transcript for the HR team. A family’s nikah recording may need only same-language captions so a hard-of-hearing grandparent can follow. These are illustrations, not rules.

What should you settle before the quote?

Each choice changes the work, so the quote should say it in writing:

• Languages: the language spoken, and each language you want on screen.
• Source of the text: written by a person, or generated by software and then corrected. The W3C notes that automatic captions need accuracy checks and usually correction. See checking automatically generated captions.
• Who reviews: someone who speaks the language and knows the names, titles and terms.
• How it is delivered: text burned into the picture, or a separate file the platform can switch on and off. This has its own guide on burned-in captions or separate caption files.
• Which videos get it: the full recording, the highlight film, the short social clips, or all of them.

The W3C also advises specifying accessibility during project planning and procurement, so it is not left to the edit. Planning accessible event videos from the brief shows how to share these tasks.

How do they differ in practice?

Captions are tied to the video’s timing, so they must be redone if the video is re-edited. A transcript is not, which is why it suits conference learning and reference, where people search for a speaker’s point.

Translated subtitles carry the most risk of misunderstanding, because the translator must choose how to render titles, company names and cultural phrases. If your audience reads Chinese or Tamil, read our notes on Chinese subtitles and Tamil and other language subtitles before you decide.

Captions and transcripts cover speech and sound. If the picture carries important information that a viewer cannot see, such as a chart or a demonstration, that is a separate question covered in video descriptions and spoken visual context.

What do we confirm with you?

• Which outputs fit your audience, and in which languages.
• Who writes the text, and who checks it before you receive the video.
• Whether captions are part of the picture, a separate file, or both.
• What we need from you, such as speaker names, a glossary or the programme.
• How corrections are handled after you review the first version.

Our caption and language planner helps you note the languages and outputs before you write to us, and the captioning and subtitles service page explains how we scope the work. Or message us your audience and languages on WhatsApp, and we will help you choose.