Synthesia lets you create a video with a realistic AI presenter without a camera, studio or microphone. You describe what you want (or paste a script), Synthesia drafts the scenes, and an AI avatar reads your text in a natural voice.

This beginner guide walks you through your first video, from a blank screen to a shareable link, using Synthesia’s built-in AI Assistant. No editing experience is needed, and you can follow along on the free plan.

Last updated: October 14, 2026. Menus and button names can change, so if something looks different, look for the closest matching option.

Time needed20–30 minutes

DifficultyBeginner

CostFree plan works

Result1–3 minute avatar video

Best for: Training, onboarding and explainer videos

Starting price: Free Basic plan; Starter $29/month

Open Synthesia Read our review

Before you start

A free Synthesia account (sign up at synthesia.io; no credit card needed for the free plan)
A topic and audience in mind, for example: “2-minute welcome video for new sales hires”
Optional: a PDF, PowerPoint, Word file or web page to base the video on
Optional: your logo and brand colours

The process at a glance

Describe
Add files
Length
Style
Outline
Polish
Share

Step-by-step: create your first Synthesia video

Step 1: Start a new video and describe it

On the Synthesia home screen, click Create video. In the prompt box, describe the topic, the audience and the goal of the video.

Example prompt: “A friendly 2-minute welcome video for new customer-support hires explaining our three values and their first-week schedule.”

The more specific your prompt (audience, tone, length, key points), the better the first draft.

Step 2: Attach your source material (optional)

Click the + (attach) button to add a PDF, PowerPoint, Word document or a URL. Synthesia uses it to write accurate scenes, which is ideal for turning an existing policy or slide deck into a video.

Attaching your own document keeps the video factual and on-message, instead of relying on the AI’s general knowledge.

Step 3: Choose the video length

Pick a Video duration: Short, Medium or Long. For a first video, choose Short; viewers finish short videos far more often.

Submit the prompt to open the setup screen.

Step 4: Pick the style and template

Under Style, choose a Delivery style:

Presentation: calm, slide-like layout, great for training and internal updates.
Dynamic: more movement and visual variety, better for marketing and social.

Set the Template to Auto or pick one from Synthesia’s library. Brand kits are available on Enterprise plans.

Step 5: Review the outline, then generate

Synthesia shows an Outline of sections. Edit any section text, delete what you do not need, or click Add to insert a section.

When it looks right, click Continue to editor to fine-tune first (recommended) or Generate video to render straight away.

Read the outline aloud. If a sentence is hard to say, it will sound awkward from the avatar too, so shorten it.

Step 6: Polish the avatar, voice and script

In the editor, click a scene to change the avatar, the voice or the language, and edit the script text directly. You can also ask the AI Assistant chat panel for changes, such as “make scene 2 shorter” or “add a call to action at the end”.

Use the preview to listen to each scene and fix pronunciation of names or product terms.

On the free plan, videos include a watermark and minutes are limited. Keep test videos short so you do not use up your allowance.

Step 7: Generate, review and share

Click Generate. Rendering usually takes a few minutes. When it finishes, watch the full video once, then share it with a link, embed it on a page, or download it as an MP4 (download options depend on your plan).

Tips for better Synthesia videos

One idea per scene. Keep each scene to 2–3 sentences and add on-screen text for key points.
Write how people talk. Use short sentences, contractions and direct questions.
Keep it under three minutes. Split longer topics into a short series.
Add visuals. Screenshots, icons and charts make an avatar video far more engaging than a talking head alone.
Translate last. Finalise the English version first, then create language versions.

Common problems and quick fixes

Problem	Quick fix
The avatar mispronounces a name	Spell it phonetically in the script (for example “Nguyen” as “Win”) and preview the scene.
The video feels too long	Delete scenes in the outline step, or ask the Assistant to shorten each scene.
The tone feels too formal	Switch the Delivery style to Dynamic and rewrite the script in a conversational tone.
Facts are generic or wrong	Attach your own document in Step 2 and edit the script before generating.

More step-by-step guides

Frequently asked questions

Is Synthesia free to use?

Yes. Synthesia has a free Basic plan with a limited number of minutes and a watermark. Paid plans start with Starter at $29 per month.

How long does it take to make a Synthesia video?

A first short video typically takes 20 to 30 minutes including editing. Rendering itself usually takes only a few minutes.

Can I use my own face as the avatar?

Yes. Synthesia offers custom avatars created from a recording of you, with a consent step. Availability depends on your plan.

Can I make the video in another language?

Yes. Synthesia supports 140+ languages. Choose a voice in the target language or translate the finished script.

Can I turn a PowerPoint into a video?

Yes. Attach the PowerPoint in Step 2 and Synthesia will build scenes from your slides.

Wrap-up

You now know the full Synthesia workflow: describe, attach, pick a style, review the outline, polish and share. Make your first video short, show it to two or three people, and use their feedback on the next one. When you are ready to compare plans or alternatives, read our full Synthesia review and our Synthesia vs HeyGen comparison.

This guide is independent and is not sponsored by Synthesia. Some links may be affiliate links, which never change our recommendations. See our Affiliate Disclosure.

Guide

How to Make an AI Avatar Video with Synthesia (Step-by-Step)