Most video caption tools make you jump through hoops — install desktop software, learn a timeline editor, manually sync every word. With CanvaSub, the entire process from sign-in to finished render takes less than 60 seconds.

We recorded the full journey on a phone to prove it. No cuts, no speed-ups, no tricks.

Here's exactly what happens in the video — and what you'll experience when you try it yourself.

Step 1: Sign In (5 seconds)

Open CanvaSub and sign in. You can use email and password, a passwordless magic link, or one-click Google sign-in. In the demo, we use Google — tap once and you're on your dashboard.

If you don't have an account yet, signing up is free and takes the same amount of time. No credit card required.

Step 2: Create a New Project (10 seconds)

From the dashboard, tap New Project. Give it a name (or let CanvaSub generate one), pick your aspect ratio — portrait 9:16 for TikTok and Reels, square 1:1 for Instagram posts, or landscape 16:9 for YouTube — and you're in.

Every project gets its own workspace where your video, transcription, captions, and exports live together.

Step 3: Upload Your Video (10 seconds)

Drop or select your video file. CanvaSub supports MP4, MOV, WebM, and most common formats. Depending on your plan, videos can be up to 30 minutes long.

The upload happens in the background — you'll see a progress bar and the video preview appears as soon as it's ready.

Step 4: AI Transcription (15 seconds)

Hit Transcribe and CanvaSub's AI engine converts speech to text with word-level timestamps. We support over 90 languages, including Arabic, Spanish, French, Hindi, Japanese, and more.

The transcription engine runs on dedicated GPU infrastructure, so even a 5-minute video transcribes in a few seconds. You'll see the full transcript appear with each word individually timed.

How accurate is the AI transcription?

For clear speech in a quiet environment, expect 95%+ accuracy out of the box. You can always edit the transcript directly — fix a name, adjust a word, and the timing updates automatically.

For details on accuracy across different conditions, see our AI transcription accuracy deep-dive.

Step 5: Choose a Caption Style (5 seconds)

Browse caption styles and pick one that fits your brand. Each style controls:

Font — family, size, weight
Colors — text, highlight, background
Animation — fade, bounce, karaoke highlight, typewriter
Position — top, center, bottom of the frame
Effects — stroke, shadow, gradient, pill background

You can preview styles in real-time on your actual video before committing. If you want to go deeper, every property is customizable — but for most creators, picking a preset and hitting render is all you need.

For inspiration, check out our guide to the best caption styles for social media.

Step 6: Render and Export (15 seconds)

Tap Render and CanvaSub composites your video with the styled captions baked in. The render runs on cloud infrastructure so it's fast — a 30-second clip typically renders in under 15 seconds.

When it's done, you get a download link for the finished video. The captions are burned into the video file itself, so they'll show up everywhere: TikTok, Instagram, YouTube, LinkedIn, Twitter/X, WhatsApp stories — anywhere you upload.

Why This Matters for Content Creators

Captions increase watch time by 40%

Social media algorithms prioritize watch time. Videos with captions keep viewers engaged longer because they can follow along even when scrolling on mute — which is how 85% of people watch social media videos.

Captions make your content accessible

Over 466 million people worldwide have disabling hearing loss. Captions aren't just a growth hack — they make your content available to everyone.

Captions boost SEO for video content

Search engines can't watch your video, but they can read your captions. Captioned videos rank higher in Google, YouTube search, and social media discovery feeds.

How CanvaSub Compares to Manual Captioning

| | Manual (CapCut, Premiere) | CanvaSub | |---|---|---| | Time per video | 15-45 minutes | Under 60 seconds | | Transcription | Type it yourself or fix auto-captions | AI with 95%+ accuracy | | Styling | Manual per-word formatting | One-click style presets | | Batch processing | One at a time | API available for automation | | Platform | Desktop app required | Works in any browser | | Mobile workflow | Limited or impossible | Full mobile support |

Frequently Asked Questions

Can I use CanvaSub on my phone?

Yes. Everything in the demo was done on a mobile phone. CanvaSub is a web app that works in any modern browser — no app install required. Upload, transcribe, style, and render all from your phone.

What video formats are supported?

MP4, MOV, WebM, AVI, MKV, and most common formats. The maximum duration depends on your plan — from 60 seconds on the free tier to 30 minutes on Team plans.

Is there a free plan?

Yes. You get free credits when you sign up, enough to try the full workflow on several videos. No credit card required to start.

Can I edit the captions after transcription?

Absolutely. The transcript is fully editable — fix words, adjust timing, add or remove segments. Changes are reflected instantly in the preview.

What languages does the AI support?

Over 90 languages, including English, Spanish, French, Arabic, Hindi, Portuguese, Japanese, Korean, Chinese, German, Italian, and many more. RTL languages like Arabic and Hebrew are fully supported with correct text direction.

Can I use my own fonts?

Caption styles come with curated font pairings, and you can customize the font family, size, weight, and spacing. Custom font upload is available on paid plans.

Try It Yourself

The best way to understand the workflow is to try it. The entire journey from creating an account to downloading your first captioned video takes under a minute.

Get started for free — no credit card, no install, no learning curve. Upload a video, let the AI do its thing, and download your captioned video.

How to Add Captions to a Video in Under 60 Seconds (Full Walkthrough)