Good cuts need words on screen: a title that says what the viewer is looking at, the selling points under the product, a price, a closing call to action. Describe them in chat and Sparki writes the wording, puts each card where it belongs in the frame and times it to your edit — no text boxes to drag, no timeline scrubbing.
You describe the cards, Sparki handles the wording, the placement and the timing.
Tell Sparki what the video needs: a title for the first three seconds, the three reasons this gadget is worth buying, the price, a booking card at the end. It drafts the wording if you have not written it, puts each card in a spot where it stays readable against the picture, and lines every card up with the shots it belongs to. Anything off — a clunky sentence, a card in the wrong corner — is one message to fix.

Text stays editable across as many rounds of revision as you need. Ask to move the price card lower, shorten the title, swap "Order now" for "Book a table", or drop the second selling point entirely — Sparki changes that one card and the rest of the edit stays as it is. Useful when a client keeps suggesting wording changes: paste their notes into the chat and send the revisions through one by one.

Text that lingers past its shot looks broken. Sparki ties each card to the timeline: the title holds for the length of the opening shot, a chapter label appears when the section starts and leaves when it ends, the call to action waits until the final frames. If a card should run longer or end sooner, say so and its timing changes without touching anything else.

Six kinds of on-screen text, each placed and timed against your footage.
Any video that opens cold — product demos, vlogs, walkthroughs, explainers — where the viewer needs one line saying what they are about to watch. Sparki drafts the title from your brief, places it where it reads cleanly against the opening shot and holds it long enough to be read before taking it off. The first three seconds, where most viewers decide whether to keep watching.
Tutorials, recipes, routines and guides that move through distinct steps a viewer may want to follow or come back to. Sparki labels each section as it starts — "Step 1: Brew the concentrate", "Part 2: Backyard setup" — and takes the label off when the next section begins. Explainers and how-tos where the structure needs to be visible on screen.
Product close-ups, demos and before-and-afters where the picture shows the feature but nobody says it out loud. Sparki writes each point in a few words, shows one card at a time and syncs it to the shot that demonstrates it, so the claim lands with the proof. Ads and organic product posts where most viewers watch with the sound off.
Menu shots, product tables, service walkthroughs — any clip where the cost or the offer is what decides the sale. Sparki gives the price or the offer its own card instead of burying it in a sentence, and keeps it up for the whole shot it applies to. Local business promos, ecommerce ads and limited-time offers.
The final seconds of any promotional video, where the viewer should do one specific thing next. Sparki writes the call to action and holds it on a clean card through the outro, keeping the wording to one instruction instead of three competing ones. Turning a view into a booking, an order or a follow.
Testimonial clips, screen recordings and talking heads where a quoted line, a definition or a disclaimer should appear on screen — text you choose, not transcribed speech. Sparki formats the line to sit like a caption — short, readable, out of the speaker's way — and times it to the moment it refers to. Stressing a spoken line or adding context without a voiceover. Speech-to-text captions are a separate capability: ask for captions and Sparki generates those from the audio.
Four people who need words on screen and no design tool.
Knows most viewers watch with the sound off, so the pitch has to be on screen. Sparki puts the three selling points over the shots that show them and the price on its own card, then trims the wording until every line reads in one glance.
Records long walkthroughs where viewers get lost halfway through. Sparki labels each section as it starts and keeps the labels consistent, so the video is easy to follow and easy to come back to mid-series.
Films the walk from the street door to the apartment once and reuses it for every booking. Sparki adds the welcome title, a card per step — lockbox, lift, parking — and a quiet closing line asking guests to message with any questions.
Walks through a property on camera and wants beds, baths, price and the viewing CTA stated in text rather than spoken. Sparki places each fact on the shot that shows it and holds a "Book a viewing" card through the outro.
Upload the cut, describe the cards, export with the text in place.
Drop in a file or paste a link, in 9:16, 1:1 or 16:9. Sparki runs in the cloud on a raw cut or on footage you have already edited, and the existing picture and audio stay untouched.
Name the cards and the exact wording, or describe what each one is for and let Sparki draft it. It places every card in a suitable spot in the frame and times each one to the shots it belongs to.
Ask for wording, position or timing changes card by card — Sparki revises each one across as many rounds as you need — then export the finished cut with the text in place.
Text overlays are one layer. Sound effects, music and style cloning work the same way, on the same upload.
Upload a video, say which titles, prices and calls to action it needs, and Sparki writes them, places them and times them — then revises any card until the wording is right.
Add text to your video for free