How do I auto-caption a long YouTube video?

Drop the video file into the tool above. Transcription speed depends on your device, and automatic recovery passes can take several times the video length. Then tune timing on the zoomable timeline and export SRT for YouTube Studio.

Is this subtitle generator really free?

Yes — unlimited use, no watermark, no sign-up. It runs Whisper AI on your device, so there are no per-minute fees. SRT, VTT, TXT, and burned-in video exports are all included.

How do I caption a TikTok video?

Upload the clip, choose a Bold preset like Comic Punch, set 1–2 words per line with the Karaoke Sweep animation, drag the caption clear of faces and buttons, then download the video with burned-in captions.

Do Reels exports have a watermark?

No. Every export — SRT, VTT, TXT, and burned-in video — is watermark-free.

Can I match captions to my channel’s style?

Yes, through styling: pick a preset family (Minimal for calm explainers, Bold for high-energy clips, Script for personal vlogs), set your colors and animation once, and reuse them on every video.

How long does a long video take?

Speed depends on the device and audio. The larger model and automatic short-window rechecks can take several times the video length. The editor opens after the first ~12 seconds are ready, and the rest streams in while you work.

How accurate are the captions?

Around 90–95% for clear English audio — usually better than YouTube auto-captions. Noisy recordings, heavy accents, and rare jargon need a quick pass in the Edit tab.

Do I need an account?

No. Open the page, drop a file, get captions. Files are processed on your device and never uploaded.

Does it work for YouTube Shorts?

Yes. Drag captions into the vertical safe zone, use 1-word pops or short phrases, and export the video with burned-in animation.

Free vs paid captioning — which should I choose?

Choose paid tools (Rev, HappyScribe) for certified transcripts, many languages, or team workflows. Choose this free editor for fast, private English captions with styling and animation.

Can I use ChatGPT with my captions?

Yes. Copy the transcript from the editor, then ask ChatGPT to write a description, summarize key points, or draft post captions with hashtags. ChatGPT cannot transcribe the audio itself.

Does it work offline?

English transcription works offline after the one-time model download (~80 MB). Style fonts load from Google Fonts, so previewing a new style for the first time needs a connection.

Can it handle multiple speakers?

Speech is captioned sequentially in one track — it does not diarize or label speakers. For interviews, add names like “ANA:” at the start of lines in the Edit tab.

Can I sell captioning services made with it?

Yes. Many freelancers draft client captions here in minutes, polish timing and style, and deliver SRT plus a captioned video. Proofread before delivery — you are responsible for accuracy.

Does it do live streams?

No — it is built for pre-recorded files, where full-context transcription is more accurate. For live shows, use your platform’s live captions, then run the recording here afterward for a clean VOD file.

Does it support non-English videos?

No. This tool transcribes English speech only; it does not detect or translate other languages. For non-English audio, use a multilingual transcription or translation tool.

How do I get professional-looking captions without an editor?

Use a clean preset (e.g. Clean Box), add a subtle outline or shadow, keep 1–3 words per caption line, pick one restrained animation like Subtle Lift, and place captions in the lower third.

SRT vs burned-in captions — what’s the difference?

SRT/VTT are closed-caption sidecar files viewers can toggle (best for YouTube accessibility). Burned-in (open) captions are part of the video pixels — best for TikTok and Reels, where most watch on mute.

What’s the fastest export path?

Transcribe, pick a style, then download SRT/VTT instantly or render the burned-in video (real-time). Text exports are immediate; video export takes as long as the clip.

YouTube auto-captions vs this editor?

YouTube’s are automatic but plain and hard to restyle. This editor gives better accuracy in most cases plus editable timing, 100+ styles, animations, and watermark-free SRT/VTT and video exports.

How do I make captions pop on Instagram?

Try Blue Pop or Neon Green Block at 30px+, two-tone accent on the last word, and a Karaoke or Bouncy Words animation. Keep captions inside the center safe zone.

Do I need to install anything?

No. It runs in Chrome, Edge, and Safari (iOS 16.4+ for the on-device engine). Video export with burned-in captions needs canvas capture, so use Chrome or Edge for that.

Does it handle accents and slang?

Common accents and everyday slang usually transcribe well since Whisper trains on diverse speech. Strong accents or niche slang may need quick corrections in the Edit tab.

How does this compare to Rev.com?

Rev sells human ($1.50/min) and AI ($0.25/min) captions. This editor is free and instant for drafts; if you need certified accuracy, draft here and proofread, or pay Rev for the final pass.

Does it remember my style?

Your last preset, colors, and size stay selected while you work, so repeat videos keep the same look. For a lasting brand kit, note your preset name and hex colors and reuse them.

What’s the catch vs $17–30/month tools?

No catch for solo creators who need English captions: unlimited minutes, styling, and watermark-free exports. Paid tools may suit you better if you need multilingual transcription, human review, or team collaboration.

How do I avoid watermarked exports?

Watermarks come from free tiers of cloud editors (Kapwing, VEED, Clideo). Everything rendered here — text or video — has no watermark, because there is no paywalled tier.

Which file formats work?

MP3, WAV, AAC, M4A, FLAC, OGG, MP4, MOV, MKV, and WebM — anything your browser can decode. On iPhone/iPad, stick to MP3, M4A, WAV, MP4, or MOV; WebM/MKV/Opus do not decode on iOS.

Can I caption many videos in a row?

Yes, one after another with no limits — finish, export, and start the next. There is no multi-file queue, so true batch happens sequentially (or across browser tabs).

TikTok caption recipe (step by step)?

1) Upload the clip. 2) Words per line: 1. 3) Preset: Retro Punch or Comic Punch. 4) Colors: white text, yellow accent. 5) Animation: Karaoke Sweep. 6) Drag to the mid-lower area. 7) Download the video and post.

How is it with medical or technical terms?

Common terms are fine; rare jargon lands around 80–85% and needs manual fixes. Speak clearly, use a good mic, then fix the handful of technical words in the Edit tab.

Best settings for Reels and Shorts?

1–3 words per line on a single caption line, a Bold or Display preset, a Karaoke or Bounce animation, and captions dragged into the vertical safe zone. Export burned-in video.

Can it do bilingual videos?

No. Transcription is English-only, so mixed-language or bilingual speech may be inaccurate. You can manually edit the transcript in the Edit tab.

Formatting tips for readable captions?

High-contrast text with outline or shadow, 26–32px size, 1–3 words per line, lower-third placement, and one animation per video. Test on a phone screen before exporting.

How much time does AI captioning save?

Manual captioning runs ~4–6× video length; this drafts in roughly 1–2× with a few minutes of review. A 30-second short is typically drafted in under a minute plus styling.

Can it handle a 2-hour podcast?

Yes. The file is transcribed in segments that stream into the editor, so you can start reviewing immediately. Long files need patience and a plugged-in device — keep the tab open.

Do captions help reach and SEO?

Yes. Most social viewers watch muted, so burned-in captions lift retention; on YouTube, uploaded SRT gives the platform accurate text to index and makes videos accessible to more people.

Can I fix mistakes easily?

Yes. The Edit tab lists every segment with timestamps — click the pencil, correct the word, press Enter. Timing is adjusted by dragging caption edges on the timeline. No re-transcription needed.

Are there any usage limits?

No minute caps, video caps, or watermarks. The only limits are your device’s memory and patience on very long files.

100% Free · 100 Animations · On-Device Private Processing

AI Caption & Subtitle Generator

Auto-generate synchronized subtitles, drag captions anywhere over your video, then style and animate them with 100+ trending presets built on 85 fonts and 100 caption animations — plus a Canva-style color palette.

Upload audio or video

Drop MP3, WAV, M4A, or MP4, MOV, WebM — private on-device processing

Audio (MP3, WAV, AAC, M4A)

Video (MP4, MOV, MKV, WebM)

English speech only · Runs directly inside your browser · No files uploaded to servers

Free AI captions for YouTube, TikTok, and Reels

LuminaAudio turns speech into synchronized subtitles right in your browser. Upload a video or audio file, get editable captions in minutes, then style them with 100+ presets and 100 animations and export SRT, VTT, TXT, or a video with burned-in captions — free, with no watermark and no sign-up. Compare it with paid options below, or explore our other free audio tools.

Updated October 2026· By the LuminaAudio team · Tested in Chrome, Edge & Safari

No WatermarkNo Sign-UpWorks OfflineBrowser-Based100+ Caption Styles100 AnimationsSRT / VTT / VideoPrivate & Free ForeverNo Limits

Add captions in three steps

No installs, no uploads to a server. A one-hour video typically takes minutes to transcribe instead of hours by hand.

01
Upload audio or video
Drop in MP3, WAV, M4A, MP4, MOV, or WebM. Files are decoded locally in your browser, so private interviews and client work never leave your device. The English speech model (~80 MB) downloads once and is cached for next time.
02
Generate and style captions
On-device Whisper AI produces word-level timestamps. Then pick from 100+ style presets across 85 fonts — Business, Retro, Bold, Elegant, Minimal, Script, and Display — and add one of 100 animations such as Karaoke Sweep, Word Slam, Mask Rise, Typewriter, or Lower Third. Adjust colors, size, words per line, and drag captions anywhere.
03
Edit and export
Fix any word in the Edit tab, fine-tune timing by dragging caption edges on the timeline, then download SRT, VTT, TXT, or a video with burned-in captions and animation. No watermark on any export.

Made for every format you publish

One editor for long videos, vertical shorts, and spoken-word audio.

YouTube videos & long-form

Caption tutorials, vlogs, and 1-hour+ uploads with precise timing you can tune on a zoomable timeline. Export SRT for YouTube Studio to support accessibility and search.

long-form · accessibility

YouTube Shorts

Keep vertical 9:16 captions inside safe zones. One-word karaoke pops or short 2-3 word phrases match the pacing viewers expect on Shorts.

vertical video · karaoke

TikTok

Pick a Bold preset such as Comic Punch or Neon Green Block, set TikTok Yellow (#FFE600), and export burned-in captions with animation. No app install, no watermark.

short-form · burned-in

Instagram Reels

Clean, readable captions with outline, shadow, or glow that survive bright footage and small screens — exported watermark-free.

reels · styling

Interviews & podcasts

Long spoken-word audio with clear word boundaries. Fix names in the Edit tab and export TXT for show notes or SRT for video versions.

spoken word · transcripts

Many videos, same workflow

Finish one video in minutes, then repeat with the same style for a consistent look. Manual captioning takes hours per hour of footage; AI drafts take minutes plus a quick review.

workflow · time saved

Why creators stick with this editor

Free forever, private by design, and honest about its limits.

No sign-up, files stay on your device

No account, email, or credit card. Audio is transcribed locally after the first model download — nothing is uploaded to a server, unlike most cloud captioning services.

Works offline after the first run

The English-only Whisper Base model runs via WebAssembly or WebGPU in your browser. The one-time download (~80 MB) is cached; transcription after that needs no internet. (Style fonts load from Google Fonts, so new style previews need a connection the first time.)

Watermark-free exports

SRT, VTT, TXT, and burned-in video exports all come out clean. Style with 100+ presets, dual-tone colors, outline, glow, and drag-to-position — what you preview is what you download.

Honest accuracy expectations

Clear English speech typically lands around 90-95% correct — better than YouTube auto-captions in most cases. Accents and slang usually work; rare jargon and names may need a quick fix in the Edit tab.

English-only transcription

The on-device Whisper Base English model transcribes English speech locally in your browser. It does not auto-detect, transcribe, or translate other languages.

A consistent look for your channel

Instead of guessing your voice, the editor lets you lock in a signature style: preset, colors, size, and animation. Reuse it across Reels and Shorts so every video looks like yours.

A step up from YouTube auto-captions

YouTube built-in captions are convenient but plain and hard to restyle. Here you get editable timing, 100+ styles, animations, and SRT/VTT plus burned-in video export.

Captions that help reach

Readable captions keep sound-off viewers watching, and SRT uploads give YouTube text to index. For post captions, paste the transcript into ChatGPT and ask for a description or hashtags.

Offer captioning to clients

Freelancers use the editor to draft client captions in minutes, polish timing and style, then deliver SRT plus a captioned video. Turnaround depends on the device; the larger speech model and recovery passes can take several times the video length.

How it compares to paid tools

Rev.com

Rev charges per minute for human ($1.50) or AI ($0.25) captions and is the right call for legal-grade transcripts. For everyday creator videos, this free editor drafts accurate captions in minutes that you can proofread yourself.

HappyScribe

HappyScribe offers 100+ languages and collaboration, but files are uploaded and free plans are limited and watermarked. This editor handles common formats locally with no minute caps.

ChatGPT workflows

ChatGPT cannot transcribe audio itself. Draft the transcript here, then ask it to tighten the wording, write a description, or suggest titles and hashtags. More how-tos on our blog.

Live streams & accessibility

Built for pre-recorded video, where full-context transcription beats live captions. Exports carry millisecond timing suitable for closed-caption (SRT/VTT) accessibility workflows.

Frequently asked questions

Short, honest answers about cost, accuracy, languages, formats, and workflows.

Ready to caption your next video?

Drop a file above and get editable, animated captions in minutes — free, watermark-free, and private. Built for YouTubers, TikTok and Reels creators, podcasters, and interviewers.

100% Free Forever100+ Styles100 AnimationsNo WatermarkWorks OfflineSRT • VTT • TXT • Video

Slowed & Reverb GeneratorSlow down + add reverb Nightcore GeneratorSpeed up & pitch shift Reverb GeneratorAdd hall & room reverb Bass BoosterBoost low-end punch Audio TrimmerCut & fade audio Audio ConverterMP3, WAV & AAC All ToolsBrowse everything

Read Our Blog