How to Make LinkedIn Videos with Synthesia

Most “AI LinkedIn video” posts skip the two facts that actually decide the project. First: a feed clip is not a product demo and it is not a YouTube episode. It is 30–90 seconds, a hook the muted scroller can read in three seconds, and one ask — usually a comment or a DM. Second: the tool this site sends for that talking-head file is Synthesia (synthesia.io). This guide is the working path for a marketer or founder who does not have a camera: script, ratio, captions, CTA, cadence.

When a LinkedIn post needs an avatar — and when it doesn't

Use an avatar when the post needs a face and that face cannot be booked this week. A weekly point of view, a launch note, a hiring note, a “here is the pattern I keep seeing in deals” clip — those are the fit. Edit the line, re-render, post. That is the economic case. The audience is other operators on LinkedIn, not a pricing-page prospect and not a YouTube subscriber.

Skip the avatar if the post is a carousel that is really a framework, a text post that is really a paragraph, or a founder letter that has to be a real face. Skip it if you are about to walk a product UI — start at the product-demo how-to and put that file on a landing page, not in the feed. Skip it if the file is an internal lesson — the training how-to is the other desk. Skip it if you are standing up a YouTube show — the faceless-channel how-to is the channel brief.

The seven-step Synthesia LinkedIn video

1. 01

   Write a 30–90 second LinkedIn take, not a pitch deck

   One idea. One audience. One next step. Open on the take in the first sentence — the line a muted feed scroller has to read in three seconds — then spend the rest of the clip proving it. Close by asking for a comment or a DM, not a pricing URL. Spell names and numbers the way they should be heard. If you wanted a sales walkthrough of the product, that is the product-demo how-to. If you wanted a learning objective, that is the training how-to. If you wanted a weekly YouTube show, that is the faceless-channel how-to.

2. 02

   Confirm a talking-head is the right LinkedIn format

   Use a presenter when the post needs a face and you cannot or will not light a founder take every time you have a point. A weekly POV, a launch note, a hiring note, or a “here is the mistake I keep seeing” clip is the fit. Skip the avatar for a carousel that is really a slide, a text post that is really a paragraph, or a founder letter that has to be a real face. Skip it if the file is a two-minute product tour — that cut belongs on a landing page, not in the feed.

3. 03

   Open Synthesia (synthesia.io) — check the domain

   Synthesia is the presenter desk this page is written for. Synthesys (synthesys.io) is a different company. HeyGen and Pictory show up in the same “AI LinkedIn video” search; they are editorial mentions here, not hops. Mixing the names is the expensive mistake. This page sends you to Synthesia only. LinkedIn itself is just linkedin.com — a normal site, not a hop.

4. 04

   Pick the plan that can hold a posting cadence

   Basic ($0) is an audition: 9 avatars, 10 minutes a month, no clean download. Starter ($29/month, or $18/month billed yearly) is where downloads and logo removal start — that is the usual feed export. Creator ($89/month, or $64/month yearly) adds minutes and personal avatars. Unused minutes do not roll over. Ten minutes a month on Starter is a handful of 45-second posts, not a daily show. Do not buy Creator hoping the avatar becomes an actor, and do not buy Enterprise for a single English clip.

5. 05

   Frame for the feed: square or vertical, captions on

   Set the project to a feed ratio before you write scenes. Square (1:1) or vertical (4:5 or 9:16) fills a phone. Widescreen 16:9 is a leftover from landing pages and YouTube. Confirm the current aspect-ratio options in Synthesia’s project settings — we do not invent a “LinkedIn preset.” Keep a head-and-shoulders crop. Turn auto-captions on; most people meet the clip muted. Do not burn a logo over the caption line, and do not leave a 16:9 letterbox in a square post.

6. 06

   Land a comment or DM CTA, then generate and QC

   The last beat is a question or an offer to continue in DMs — “Comment ‘deck’ and I will send the one-pager,” or “DM me the bottleneck.” Say that line out loud and put it on a card. Do not stack a calendar link, a blog URL, and a newsletter on the same end frame. Generate, then watch the first three seconds muted on a phone. Mouth-sync is consistently strong in English and major European languages. Hand gestures loop. These are narrators. Edit the script and re-render instead of re-booking a studio.

7. 07

   Export and post native on LinkedIn on a cadence you can hold

   Export on the plan that grants a clean download. Upload the file in the LinkedIn composer at linkedin.com — native video, not a YouTube link sitting in a text post. Write the caption as the first sentence of the take, then the CTA. If the current composer asks you to label AI-generated or synthetic media, use that control; check the live composer, not a blog post from last year. That label is separate from Synthesia’s vendor license. Pick a cadence the minute meter can actually hold, then keep it. Confirm commercial terms on the plan you will publish from.

Talking-head vs founder camera vs a carousel

“AI LinkedIn video tool” is a search, not a product. The decision is the format. Pricing on the Synthesia column matches the live synthesia.io/pricing page as of August 2026. The yearly toggle shows Starter at $18/month; the FAQ on the same page still quotes $264/year. We report both. Unused minutes do not roll over.

The script: hook in three seconds, one idea, one ask

Write the clip the way you would brief a founder for a 60-second take, then cut the greeting. “Hey everyone, I wanted to share…” is three seconds of nothing. The first sentence is the take. The next 20–70 seconds prove it with one example. The last sentence is the ask. Most feed clips should land between 30 and 90 seconds. Longer than that and you are making a demo or a webinar.

The ask on LinkedIn is usually a comment or a DM. “Comment ‘template’ and I will send it.” “What is the bottleneck on your side — reply and I will answer.” A URL on the end card is a landing-page move; if you need a link, put it in the first comment after you post, not as the only reason the file exists. Spell company names and numbers the way they should be said. If Legal has not signed the claim, do not put it in the avatar’s mouth.

Do not write a learning objective. Do not write a feature tour. Do not ask the avatar to “sound excited.” These presenters are narrators: blink cadence, micro head-tilt, a looping hand. On a head-and-shoulders framing that is enough. Full-body action and emotional range are not the product. Full breakdown of what the avatars actually do is in our Synthesia review.

Square or vertical, captions on, native upload

The picture job is the feed, not a hero embed. Set square (1:1) or vertical (4:5 or 9:16) before you generate. Widescreen 16:9 is what you export for a website. Confirm the current aspect-ratio options in the project — we do not invent a LinkedIn button inside Synthesia. Keep the crop tight: head-and-shoulders, a background you can reuse, no chrome from a product UI unless that UI is the point (and then you are probably on the demo brief).

Captions go on so the file works muted. Auto-captions are the start; fix names and numbers before you export. Brand kits — locked colors, logos, fonts — sit on Enterprise. For a first LinkedIn clip, a stock presenter, a clean Starter download, and burned-in captions are the whole buy. Upload the file in the LinkedIn composer. A YouTube link in a text post is not a native video, and it is not this workflow.

Which Synthesia plan ships the LinkedIn clip in 2026

The 2026 pricing page meters Synthesia in credits plus video minutes. Basic is $0 with 9 avatars and 10 minutes a month — downloads and logo removal start on Starter. Starter is $29/month (1,200 credits, 10 minutes of video or dubbing, 125+ avatars) or $18/month billed yearly. Creator is $89/month (3,600 credits, 30 minutes, five personal avatars, API) or $64/month yearly. Annual Starter lists 14,500 credits and 120 minutes a year; annual Creator lists 44,000 credits and 360 minutes a year. Languages and voices are published as 160+ on every tier. Unused minutes do not roll over.

For a first feed clip, start on Basic with a real take — the claim you would actually post, not a placeholder sentence. Go to Starter when you need a logo-off file. Buy Creator for a personal founder avatar or the API — not for “more cinema.” Buy Enterprise when brand asks for a kit or you are ready to localize a locked cut. Confirm commercial terms on the plan you will export from. Ten minutes a month is a small stack of 45-second posts, not an always-on channel.

The comment or DM close — then a cadence

The last scene is the reason the file exists. Put one ask on a card and have the avatar say that same line. A comment keyword you can search later. A DM you will actually answer. Not three next steps. Not a fade to black and a hope that the caption does the work. If the post is a hiring note, the ask is “comment ‘roles’.” If it is a POV, the ask is a question the right people can answer in one sentence.

Watch the export the way the feed will: on a phone, muted, then with sound. The first three seconds have to name the take. The last three have to name the ask. Proper nouns are where avatar reads go wrong — fix the script and generate again. That loop is the product. A studio re-book is what you are not buying.

Then pick a cadence and keep it. We do not print impression or follower targets. Consistency is the operator move; volume is how you burn the minute meter. Starter’s 10 minutes a month will not fund a daily 90-second show. A few tight clips a week will. Schedule in the LinkedIn composer. If the current upload flow asks you to label synthetic or AI-generated media, do that — it is separate from the vendor license, and you need both when the clip is public.

Localization and founder likeness are next steps, not this brief

Ship the home-market clip first. A second language is a translation of a signed take, not a reason to start in Enterprise. Synthesia’s 1-click translation into 80+ languages and brand kits live on that tier. When you are ready, lock the English (or home-market) file, then duplicate. The working path for a locked training master is how to localize training videos — same company, different job. A feed clip you are still rewriting does not get a Spanish twin yet.

A custom founder avatar is also a next step. Prove the format on a stock presenter. Then decide whether the audience needs that face. The comparison of presenter platforms — including where Synthesys sits as a different company — lives on Synthesia vs Synthesys and Best AI Avatar Video Tools 2026. This page does not hop there. The only Try button here is Synthesia.

HeyGen, Pictory, and a phone camera — editorial only

Those names sit in the same “AI LinkedIn video” search. HeyGen is the usual Synthesia alternative in reviewer write-ups: faster custom-avatar turnaround, punchier social presenters, less of the enterprise kit. Pictory is closer to a text-to-video editor than a governed presenter desk. A phone camera is the right tool when the founder has to be the one on camera and you can actually book the take. None of them have a live hop on this page, so there is no Try button. If you need the talking-head feed clip we can route, it is Synthesia.