Closed Captions vs Subtitles: Differences, Hard vs Soft Subs

The difference between closed captions vs subtitles comes down to who each is written for. Closed captions assume the viewer can't hear the audio. They include the dialogue plus speaker labels and meaningful sounds such as [phone ringing] or [tense music], and the viewer can switch them on or off. Subtitles assume the viewer can hear but doesn't understand the spoken language, or simply prefers to read along. They usually contain only the dialogue, often translated. A related distinction is soft vs hard subs. Soft subtitles are a separate track that viewers can turn off. Hard (burned-in) subtitles are drawn into the picture permanently. For accessibility, you need captions. For a foreign-language audience, you need subtitles. For social feeds where video autoplays muted, burned-in captions are often the most reliable way to get your words seen.

Key takeaways

• Captions = dialogue + speaker identification + important sounds, for viewers who are deaf or hard of hearing. Subtitles = dialogue only, usually for language access.
• "Closed" means the viewer can toggle them. "Open" captions are always on, because they're burned into the video.
• SDH (subtitles for the deaf and hard of hearing) is a subtitle file written with caption-style sound cues. It's handy where a platform doesn't offer a separate captions track type.
• Soft subs are flexible, searchable and multi-language. Hard subs show up everywhere but can't be changed, translated or turned off.
• WCAG 2.2 requires captions for prerecorded video at Level A and for live video at Level AA. In the US, the ADA Title II rule points public entities to WCAG 2.1 Level AA.

Closed captions vs subtitles: the core difference

People often use the two words interchangeably, especially outside North America, where "subtitles" often covers both. In US practice, and in web standards, they mean different things. The HTML specification, for example, has separate captions and subtitles track types. MDN describes captions as dialogue plus sound effects and other audio cues for when sound is unavailable or unclear, and subtitles as a transcription or translation of the dialogue for when the sound is available but not understood.

The simplest test: if you muted the video, would the text tell you everything you need? If yes, it's a caption track. If it only tells you what people said, it's a subtitle track.

What closed captions include

Good captions carry everything in the audio that a viewer needs to follow the content. W3C's guidance for WCAG Success Criterion 1.2.2 says captions identify who is speaking and describe meaningful non-speech audio, such as sound effects and music that carry meaning. Here is a short captioned scene in SRT format:

1
00:00:00,500 --> 00:00:02,400
[rain against the window]

2
00:00:02,800 --> 00:00:05,100
JORDAN: Did you lock the back door?

3
00:00:05,600 --> 00:00:07,900
(from hallway) I thought you did.

4
00:00:08,300 --> 00:00:10,000
[glass shatters]
[ominous music]

Notice the conventions. Square brackets mark sounds. A speaker name appears when it isn't clear who's talking, such as an off-screen voice or a new character. A parenthetical note says where an off-screen voice comes from. Caption every sound that matters to the story or the information, but skip ones that don't. A dog barking in a podcast recording usually isn't worth a caption. A dog barking in a thriller just before the door creaks probably is.

What subtitles include

Subtitles give a viewer who can hear the audio access to the words, usually because the words are in another language. A Spanish-language documentary with English subtitles shows the English translation of the dialogue and nothing else. Sound effects are left out, because the viewer can hear them.

Same-language subtitles are common too. Many people watch with them on because of accents, poor audio, noisy rooms or simply preference, and language learners use them to match spoken and written words. Same-language subtitles still aren't full captions unless they include the speaker identification and sound cues a viewer who can't hear would need.

What SDH means (subtitles for the deaf and hard of hearing)

SDH is a hybrid. It uses the delivery of a subtitle track with the content of captions: dialogue, speaker labels and sound descriptions. SDH emerged because some formats and players, such as Blu-ray discs and many streaming apps, only had a "subtitles" track type. Distributors put caption-style content into that slot and labeled it SDH.

On the web, you have a real choice. The HTML track element has both kind="captions" and kind="subtitles", so a file with sound cues belongs in a captions track. On platforms that only take a subtitle file, an SDH-style file is the way to make that file accessible.

Open captions vs closed captions

The "closed" in closed captions means the captions are hidden until the viewer turns them on. Open captions are part of the picture and are always visible. WCAG's glossary recognizes both. Closed captions can be turned on and off with some players, and open captions can't be turned off. Both can satisfy the captioning requirement if they're accurate and complete.

So the question isn't which one is "accessible." It's which one fits where the video will play. Closed captions let viewers choose and let you offer several languages. Open captions guarantee the text is seen, even in players that ignore caption tracks.

Hard sub vs soft sub: pros and cons

"Soft" and "hard" describe how subtitles are delivered, whether they contain captions or translated dialogue.

• Soft subtitles are a separate stream. It can be a sidecar file uploaded next to the video, such as an SRT or VTT on YouTube, or a subtitle track stored inside the video container, such as MP4 or MKV. The player draws the text, so the viewer can switch it on or off, pick a language and often adjust the size and color.
• Hard subtitles (burned-in, also called open captions or hardcoded subtitles) are rendered into the video frames. Every player shows them, and nobody can remove them.

When to burn subtitles in

Burn in when you can't control the player. Short-form social feeds often autoplay video muted, and a clip that gets downloaded and reshared through messaging apps loses any separate subtitle file along the way. Burned-in captions are also the practical choice for:

• Video ads and promos, where the first seconds have to work with the sound off.
• Event screens, kiosks and presentations, where you don't know what software will play the file.
• Stylized "talking head" clips, where the text is part of the visual design.

When to keep them soft

Keep subtitles soft when you publish on platforms that support caption tracks, need more than one language, expect to correct the text later, or want viewers to control the appearance. Soft tracks also keep the original picture untouched, which matters for archival masters. A good habit is to keep a soft master with a clean subtitle file, then burn captions into the social cut-downs only.

Accessibility standards and laws you should know

This section is general information, not legal advice. Requirements depend on who you are and where you publish.

WCAG 2.2

The W3C's Web Content Accessibility Guidelines are the reference most accessibility laws and policies point to.

• 1.2.2 Captions (Prerecorded), Level A. Captions are provided for all prerecorded audio content in synchronized media, unless the media is itself a clearly labeled alternative to text.
• 1.2.4 Captions (Live), Level AA. Captions are provided for all live audio content in synchronized media.

Because 1.2.2 is Level A, it's a baseline requirement. A site can't claim any WCAG conformance level while its prerecorded videos lack captions. Translated subtitles without sound cues don't meet the criterion on their own.

FCC caption quality standards (US television)

For programming shown on US television, the FCC's rules at 47 CFR 79.1(j)(2) set four quality standards:

1. Accuracy. Captions match the spoken words, in the original language.
2. Synchronicity. Captions line up with the spoken words and sounds as closely as possible, and stay on screen long enough to read.
3. Completeness. Captions run from the beginning to the end of the program.
4. Placement. Captions don't block important visual content.

Even if your video never airs on TV, those four standards make a good quality checklist for any caption file.

ADA

In the United States, the Department of Justice's 2024 rule under ADA Title II requires state and local government websites and mobile apps, including their video content, to meet WCAG 2.1 Level AA. In April 2026, the DOJ issued an interim final rule that pushed the compliance dates back by about a year: to April 26, 2027 for entities serving 50,000 or more people, and to April 26, 2028 for smaller entities and special districts. The technical standard stayed the same. Because these dates have already changed once, confirm the current deadlines on ADA.gov before you plan around them. Businesses covered by Title III don't have an equivalent technical rule, but captioning is widely treated as part of making video accessible.

How to insert subtitles in a video

Whichever delivery you choose, you first need a subtitle or caption file. If you don't have one, generate it from the audio. Then choose how to attach it.

Step 1: Create the subtitle file

1. Open the subtitle generator and sign in with your free WaveToWord account.
2. Upload your video or audio, choose the spoken language, and pick SRT or VTT.
3. Download the file and proofread it. Correct names and technical terms, and for captions, add sound cues such as [applause] and speaker labels where needed. The browser-based subtitle editor lets you fix lines, adjust timings, shift everything at once and preview against your video without uploading anything.

The generator handles media up to 15 minutes per run, with two free AI runs per tool per day. For longer material, upload it to the Files library in your free account, which is not limited to 15 minutes and exports SRT or VTT from the finished transcript. If you're unsure which file format to use, our guide to SRT vs VTT subtitle files explains the trade-offs.

Step 2a: Add a soft subtitle track with WaveToWord

1. Open add subtitles to video and sign in.
2. Upload the video (MP4, MOV, MKV, WebM, AVI, M4V, WMV or FLV) and the subtitle file (SRT, VTT, ASS/SSA or SBV).
3. Choose the subtitle language so players can label the track correctly, then choose the output container: MP4 or MKV.
4. Download the result. Viewers can turn the track on or off in any player that supports subtitle tracks.

The tool copies the picture without re-encoding it whenever the container allows, so quality doesn't change and processing is fast. In MP4, subtitles are stored in MP4's timed-text format. In MKV, they're kept as SRT. If the video uses a codec that MP4 can't carry, the tool converts the picture to H.264 so the file plays broadly. If you want to keep the original codec, choose MKV. One caveat: many web browsers don't display subtitle tracks embedded inside a video file. For a website, serve a separate VTT file instead, as shown below.

Step 2b: Burn subtitles in with WaveToWord

1. Open burn subtitles into video and sign in.
2. Upload the video and the subtitle file.
3. Choose a font size (small, medium or large) and a position (bottom or top). Pick top if the bottom of the frame holds names, prices or platform buttons.
4. Download the MP4. The captions appear as white text on a semi-transparent dark box, which stays readable on bright and busy footage.

Burning requires re-encoding the picture. The tool keeps the bitrate close to the original so the file doesn't balloon in size. Because the captions become part of the picture, do your proofreading before you burn. Fixing a typo afterward means exporting again.

Step 2c: Upload a caption file to YouTube

1. In YouTube Studio, open Content, select the video and go to Subtitles.
2. Add the language if it isn't listed, then choose to add subtitles or captions and pick Upload file.
3. Select With timing, upload your SRT or VTT file and publish.

YouTube accepts many caption formats, including SRT, SBV, VTT and TTML. Its help center says that SRT style markup isn't recognized, so keep SRT files plain. If you need captions from an existing YouTube video rather than for one, see how to get a transcript of a YouTube video.

Step 2d: Add captions to a video on your own website

Browsers load subtitle text through the HTML track element, which reads WebVTT files:

<video controls src="/media/tour.mp4">
  <track kind="captions" src="/media/tour.en.vtt" srclang="en" label="English (CC)" default>
  <track kind="subtitles" src="/media/tour.fr.vtt" srclang="fr" label="Français">
</video>

Use kind="captions" for the file with sound cues and kind="subtitles" for translations. If the VTT files live on a different domain from your page, the video element needs a crossorigin attribute and the file server must allow cross-origin requests.

Alternative: do it yourself with ffmpeg

If you work on the command line, ffmpeg can do both jobs. The first command adds a soft track to an MP4. The second burns subtitles into the picture, which needs an ffmpeg build with libass:

# Soft track: copy video and audio, convert subtitles to MP4 timed text
ffmpeg -i input.mp4 -i captions.srt -map 0:v -map 0:a? -map 1:0 \
  -c:v copy -c:a copy -c:s mov_text -metadata:s:s:0 language=eng output.mp4

# Hard subs: draw subtitles into the frames (re-encodes the video)
ffmpeg -i input.mp4 -vf "subtitles=captions.srt" -c:a copy output-captioned.mp4

Desktop players such as VLC also load a subtitle file that sits next to the video with the same file name, which is handy for quick reviews before you export anything.

A quality checklist for captions and subtitles

Use this before you publish, whatever the delivery method:

• Accurate. Proper nouns, numbers and technical terms are spelled correctly. Read the file while listening, not just skimming.
• In sync. Each line appears when it's spoken and stays long enough to read. Early or late captions are distracting, and drift is worse.
• Complete. No gaps at the start, end or during music breaks. If there's meaningful sound, there's a caption for it.
• Well placed. Captions don't cover faces, on-screen text or lower thirds. Move them to the top for scenes with graphics at the bottom.
• Readable. Keep it to one or two lines per caption, break lines at natural phrases, and allow enough display time. Many style guides aim for roughly 32 to 42 characters per line.
• Labeled. Soft tracks have the correct language code, so players list them properly.

Common mistakes and how to fix them

Honest limitations

WaveToWord's subtitle tools cover the common workflows, but you should know where they stop.

• Styling isn't carried over. When you add or burn subtitles, the text is imported as plain cues. Italics, colors and positions from an ASS or styled VTT file aren't kept. Burned captions use one clean style, with your choice of size and top or bottom placement.
• One subtitle track per file. The add-subtitles tool attaches a single language track, and running it again on its own output replaces that track rather than adding a second one. For multilingual releases, upload separate subtitle files to the platform, or mux several tracks yourself with ffmpeg.
• Automatic captions are a draft. The subtitle generator transcribes the speech. It doesn't automatically write sound descriptions or speaker names in caption style. Add those yourself for accessibility-grade captions.
• Burning is permanent and re-encodes. It's the right choice for social clips, not for archival masters.

Insert subtitles into your video now

If your video is headed for a social feed, burn the captions in so everyone sees them, even with the sound off. If it's going to YouTube, your website or a streaming platform, attach a soft track so viewers can choose. Start with add subtitles to video for a switchable track, or burn subtitles into video for always-on captions. Both are free with a WaveToWord account. If you don't have a subtitle file yet, the subtitle generator will create one from your video's audio.