Social Video Transcription and Accessibility by Platform

Platform-by-platform caption support

Social platforms vary significantly in how they handle video captions, from full SRT upload support with a built-in editor (YouTube) to no external caption support at all (Instagram standard posts, TikTok). The table below shows the current state as of 2026; platform policies change and should be verified for specific use cases.

YouTube: the most complete caption infrastructure

SRT upload. Caption editor. Auto-captions. Full infrastructure. YouTube has the most complete caption support of any major platform. By a significant margin.

YouTube accepts SRT and VTT file uploads through YouTube Studio (Subtitles section). It auto-generates captions for all public videos and provides a text-based editor to correct them. Auto-caption quality ranges from 88-93% for standard English in clean recording conditions, dropping for accents, technical vocabulary, or noisy audio. For content required to meet WCAG 2.1 Level AA compliance, auto-captions alone are not sufficient — upload a corrected SRT file for any content that carries a legal obligation.

Facebook: SRT upload with filename requirements

Facebook accepts SRT uploads in Creator Studio. Language code required in the filename. Auto-captions exist but are not accurate enough for formal content. Upload the reviewed SRT. Do not rely on auto-captions for anything that matters.

Facebook accepts SRT caption files via Creator Studio for standard video posts. The filename must include a language code: video.en_US.srtfor US English. Files without the language code are rejected without a clear error message. VTT is not accepted. The SRT comma timestamp separator is required (not period).

For step-by-step Facebook captioning, the subtitle upload guide covers the exact process including common rejection causes.

Instagram and TikTok: the burned-in fallback

No SRT upload. No VTT upload. Neither platform accepts caption files. Burn captions in before upload, or use the platform's auto-caption tool. Two options. That is all.

Instagram standard posts do not support caption file uploads. Reels have limited auto-caption support with an in-app editor. For reliable, accessible captioning on Instagram, burned-in captions are the only consistent method: encode the captions directly into the video file before uploading.

TikTok has auto-caption generation for most content with a post-upload editor. The quality and editing interface has improved significantly since 2022, but it still does not accept external caption files. For creators who need accurate captions without manual editing in the TikTok app, pre-encoding with CapCut (the native TikTok editing tool) is the most efficient path.

Legal requirements and accessibility compliance across platforms

In the United States, the Americans with Disabilities Act (ADA) requires organisations to make video content accessible when the content is a service or programme that is required to be accessible under Title II (government entities) or Title III (public accommodations). The WCAG 2.1 Level AA standard, specifically Success Criterion 1.2.2, requires synchronised captions for all pre-recorded audio-visual content. Auto-generated captions on any current platform do not meet this standard without human correction.

In the UK, the Equality Act 2010 and the Public Sector Bodies (Websites and Mobile Applications) Accessibility Regulations 2018 impose similar obligations on public bodies. The EU Web Accessibility Directive applies to public sector bodies across member states. For private organisations, the business case for captions is more often engagement-led than compliance-led — but the compliance obligation for larger organisations and educational institutions is real and enforceable.

The practical implication: auto-captions satisfy accessibility requirements for informal content but not for training material, educational content, official communications, or any video that is part of a product or service. A 15-minute onboarding video published by a company is not in the same regulatory category as a casual social media post. Understanding which category your content falls into determines how much review effort captions genuinely require.

Burned-in captions: when to use them and how

Burned-in captions (also called open or hardcoded captions) are encoded permanently into the video pixels. They display on every platform, in every player, with no user settings required. They are the only reliable way to ensure captions appear on platforms that don't support caption file uploads — Instagram standard posts and Twitter/X native video are the two platforms where burned-in is the only option for captioned content.

The downside: once burned in, captions can't be corrected or localised without re-exporting the video. The production workflow: export the transcript as SRT, import the SRT into a video editor (CapCut, Premiere Pro, and DaVinci Resolve all support SRT subtitle import), position and style the captions, export the video. The exported file displays captions everywhere without any additional upload step.

For creators publishing short-form content across multiple platforms, a single captioned export covers Instagram, TikTok, Twitter/X, and any other platform without caption file support simultaneously. The workflow: transcribe once (5 minutes), review the SRT for accuracy (5 minutes), burn the captions into the video render (5 minutes including export time). Approximately 15 minutes per video for full cross-platform caption coverage with no per-platform repetition.