SDH stands for Subtitles for the Deaf and Hard of hearing. It is a specific subtitle style that includes dialogue plus non-speech sound information, like [phone rings] or [door slams], and sometimes speaker labels when it is not obvious who is talking. SDH sits in the gap between plain subtitles and traditional closed captions, and streaming platforms increasingly ask for it by name.
SDH vs subtitles vs closed captions
Dialogue text
- Plain subtitles
- Yes
- Closed captions
- Yes
- SDH
- Yes
Sound effect tags
- Plain subtitles
- Usually no
- Closed captions
- Yes
- SDH
- Yes
Speaker labels
- Plain subtitles
- Rarely
- Closed captions
- Sometimes
- SDH
- Usually
Built for
- Plain subtitles
- Language translation
- Closed captions
- Accessibility (broadcast standards)
- SDH
- Accessibility, delivered like subtitles
Where you'll see it required
- Plain subtitles
- International releases
- Closed captions
- Broadcast TV
- SDH
- Streaming platforms (Netflix, etc.)
The short version: SDH is what happens when a streaming platform wants accessibility-grade information (sound cues, speaker labels) but wants to deliver it through the subtitle track rather than a separate broadcast-style closed caption track. Broader context on the plain subtitles vs captions split: closed captions vs subtitles explained.
Why SDH exists as its own category
Traditional closed captions were built around broadcast television standards, with specific formatting and delivery rules for TV signals. Streaming platforms deliver everything as a subtitle track technically, but still want the deaf and hard-of-hearing accessibility information that closed captions historically carried. SDH is the answer: subtitle-track delivery, closed-caption-level content.
That is also why Netflix and similar platforms often list "SDH" as a distinct subtitle language option, separate from a plain "English" subtitle track for the same title.
What actually goes into an SDH file
An SDH file typically includes:
- Full dialogue, same as any subtitle track
- Non-speech sound cues relevant to understanding the scene: [tense music], [gunshot], [crowd cheering]
- Speaker identification when the speaker is off-screen or unclear: JOHN: or (over radio)
- Sometimes tone or emphasis cues where they change meaning
What it usually does not include: full sound-design detail for its own sake. The goal is comprehension, not a literal transcript of every audio event.
Do you need SDH, or will regular captions do?
Ask these questions before you build sound-effect tags into a file:
- Is this going to a streaming platform with a specific delivery spec? If a platform's style guide explicitly asks for SDH, follow their format exactly, they often have precise formatting rules.
- Is accessibility the actual goal, or just "captions so people can watch on mute"? Most muted-autoplay viewing on social media does not need sound-effect tags. Dialogue-only SRT/VTT is enough.
- Do you have a compliance requirement (ADA, broadcast regulation)? Then check the specific standard rather than assuming SDH-style tagging automatically satisfies it. Requirements vary by region and platform. The W3C's captions and subtitles overview is a reasonable starting reference for the accessibility framing.
For everyday YouTube uploads, client video, or social content, plain dialogue captions in SRT or VTT cover the job without SDH-style tagging.
How to build an SDH-style file from a transcript
- Upload your audio or video and get a timed transcript with speaker labels.
- Clean up names and dialogue accuracy as you would for any caption file.
- Manually add sound-effect tags at the relevant timestamps. AI transcription tools transcribe speech; they do not automatically detect and tag ambient sound events.
- Add or confirm speaker labels for any line where the speaker is off-screen or unclear.
- Export as SRT or VTT and format tags to match the platform's specific SDH style guide if one exists.
File Transcribe handles steps 1 through 2 and the SRT/VTT export directly: upload the file, get speaker-labeled segments, fix dialogue, export after signing in free. The sound-effect tagging step for full SDH compliance is a manual add on top, since that is a sound-design task, not a speech-to-text one.
FAQ
What does SDH mean in subtitles?
Subtitles for the Deaf and Hard of hearing. It is a subtitle style that adds sound-effect descriptions and speaker labels on top of regular dialogue subtitles.
Is SDH the same as closed captions?
Close, but not identical. Closed captions are historically tied to broadcast delivery standards. SDH carries similar accessibility content (sound cues, speaker labels) but is delivered as a subtitle track, which is how streaming platforms distribute it.
Do I need SDH for a YouTube video?
Usually not. YouTube's caption system does not require SDH-specific formatting. Dialogue-accurate SRT or VTT is sufficient unless you have a specific accessibility requirement asking for sound-effect tags.
Can AI transcription generate SDH automatically?
AI transcription tools transcribe spoken dialogue accurately, which covers the text portion of SDH. Sound-effect tags and speaker labels for off-screen speakers typically require a manual pass, since most models are not trained to detect and describe ambient sound events.
What is the difference between SDH and regular subtitles?
Regular subtitles usually contain dialogue only, aimed at viewers who don't understand the spoken language. SDH adds sound-effect and speaker information aimed at viewers who cannot hear the audio at all.
Related reading
For the broader captions vocabulary, start with closed captions vs subtitles explained. For the file format question, see what is a VTT file.
Related: Closed captions vs subtitles explained · What is a VTT file · Subtitle generator · Try File Transcribe
More guides
- Transcription guides
- Try File Transcribe free
- Transcript format guide
- Best transcription software
- Transcribe YouTube videos