Synthesia Quick Start

Learning Objectives

Understand what Synthesia is and what it produces
Navigate the dashboard and find every feature you need for your first video
Explain the scene-based structure so you can plan your video before you build it

What Synthesia Is — And What It Is Not

Synthesia is an AI video generation platform. You type a script. You choose an avatar — a digital human presenter. You design what appears on screen alongside the avatar. You click Generate, and Synthesia renders a video — complete with lip-synced narration — without a camera, microphone, recording studio, or video editor.

The result looks like a professionally produced training video. Because it is one.

What It Is Good For

Training and onboarding videos your team will actually watch
Process walkthroughs and SOPs that used to live as text documents no one read
Compliance content that needs to be consistent, documented, and easy to update
Software demos combined with screen recordings
Content that needs regular updates — because editing is fast and cheap

What It Is Not Good For

Live, real-time content that changes daily
Moments where learners need to see a specific real person — a company founder, a patient, a key client
Highly interactive, branching simulations (use Articulate or similar for those)

💡 Pro Tip

If you are hesitating over whether to use Synthesia for something, ask yourself: "Would a recorded person be better — or just more familiar?" Most of the time, it would just be more familiar. Familiar and better are not the same thing.

Dashboard Orientation

Area	What It Does	Your Priority
My Videos	Your full video library — all projects live here	High — create a folder before your first video
Templates	Pre-built video structures to study or adapt	Medium — browse one, then build your own
Media Library	Storage for logos, images, screen recordings	High — upload your logo before building
Brand Kit	Colors, fonts, logo applied automatically	High — set this up in Module 4
New Video	Creates a new project	High — you will click this in Module 3

📌 Note

Before you build your first video, create a folder in My Videos. Go to My Videos → New Folder. Name it something like "Training Videos — 2026." An organized library from day one saves significant time later.

The Scene — The Most Important Concept in Synthesia

Every Synthesia video is made up of scenes. A scene is the basic unit — like a single slide in a presentation, except it also plays narrated audio. Every scene has three parts:

Script — the words the avatar speaks. You type this.
Layout — the visual structure. The avatar's position, background, any text or images on screen.
Avatar and voice — who delivers the narration and how it sounds.

When you click Generate, Synthesia processes every scene — synthesizing the voice, animating lip movements, compositing visuals — and stitches them into a single video file.

How Long Is a Scene?

Words in Scene	Approximate Duration	Best For
50 words	~23 seconds	Intro, hook, quick transitions
75 words	~35 seconds	Most instructional content scenes
100 words	~45 seconds	Complex steps or detailed explanations
150+ words	70+ seconds	Getting long — consider splitting into two scenes

A 5-scene video where each scene is about 75 words runs approximately 3–4 minutes — a solid, complete training video for most topics.

🏁 Quick Win

Your key takeaway: Synthesia = scenes. Each scene = one script + one layout + one avatar/voice. Plan your video as a list of 5 scenes before you open the editor. This single habit separates learners who feel in control from those who feel lost.

Module 1 — Check Your Thinking

1. Which of the following is Synthesia best suited for?

A live town-hall video where the CEO addresses the company in real time A compliance training series on data security policy that is updated annually A branching simulation where learner choices change the scenario path

2. A 5-scene training video where each scene is roughly 75 words will run approximately how long?

8–10 minutes 3–4 minutes 1–2 minutes

3. What are the three parts of every Synthesia scene?

Background, music, and animation Script, layout, and avatar/voice Intro, body, and conclusion

Learning Objectives

Understand why document text never sounds right when spoken by an AI avatar
Write a script in spoken English — not written English
Use the five-scene structure to plan your first video
Draft a complete 5-scene script using the provided templates

The One Rule That Changes Everything

Write for the ear, not the eye.

Written language and spoken language are completely different. When you read, your eye can scan ahead, re-read a confusing sentence, and slow down at complex parts. When you listen, you get one pass. No scroll bar. No rewind. The listener either keeps up or falls behind — and if they fall behind, they stop paying attention.

This is why copying text from a document into Synthesia never works. A paragraph that reads perfectly on paper sounds robotic, dense, and exhausting when an AI avatar delivers it at a fixed pace.

The Five Rules of Spoken English for Synthesia

Short sentences. If a sentence is longer than 20 words, split it into two.
Active voice. "The manager approves the request" — not "The request is approved by the manager."
Use contractions. "You'll learn this in step two" — not "You will learn this in step two." Contractions sound human.
No visual references. Never say "as you can see here" or "click the button below." Describe explicitly: "Click the blue Submit button in the lower right corner of the screen."
Read it aloud. Every script. Before you build. If you stumble, rewrite.

⚠️ Watch Out

The most common mistake: writing the script last. The script is the foundation — build it first, approve it before you open Synthesia. A script change after generation costs 10 minutes. A script written well before generation costs nothing extra.

The Five-Scene Structure

Your first video has five scenes. Use this structure for any training topic:

Scene	Name	Purpose	Length
1	The Hook	Name the problem or outcome. Earn the next 4 minutes. Do NOT open with "Welcome to this training..."	40–60 words
2	The Context	Give learners what they need to know before the how-to begins	60–80 words
3	The How-To	The core process — step by step, specific and sequential	75–100 words
4	The Watch-Outs	Common mistakes, exceptions, and what to do when things go sideways	50–75 words
5	The Landing	One key takeaway, next step, and where to go for help	40–60 words

🎯 Real Talk

This structure works for almost every short training video. It is not the only structure — but it is the right starting point. Once you have built five videos using it, you will naturally adapt it. Until then, follow it exactly.

Script Templates — Fill In the Brackets

🎬 Scene 1 — The Hook

"[Name the situation or problem your learner faces]. In the next [X] minutes, you'll know exactly how to [specific skill or outcome] — [brief why it matters to them personally]. Let's get into it."

EXAMPLE: "A patient calls to schedule their first appointment. You have about 90 seconds to ask the right questions and book them correctly — or create a problem that takes twice as long to fix. In this video, you'll learn the exact sequence that makes every new patient intake smooth, fast, and accurate."

🎬 Scene 2 — The Context

"Before we get into the steps, here's what you need to know. [1–3 sentences of essential background — system used, who is involved, what triggers the process, any relevant policy]. Got it? Let's walk through it."

🎬 Scene 3 — The How-To

"Here's the process. Step one: [action]. Step two: [action — and any important detail]. Step three: [action]. Step four: [action]. Once you've completed step four, [what happens next or what confirmation to expect]."

🎬 Scene 4 — The Watch-Outs

"A few things to watch out for. [Common mistake 1] — when this happens, [what to do instead]. [Common mistake 2 or exception] — in this case, [how to handle it]. If you're ever unsure, [where to go or who to ask]. Better to pause and check than to push through and create a bigger problem."

🎬 Scene 5 — The Landing

"You've got it. The most important thing to remember: [one-sentence key takeaway]. Your next step is [specific action]. If you run into something this video didn't cover, [resource or contact]. You're ready."

Controlling Pacing With Punctuation

Synthesia's voice engine reads your punctuation as performance instructions:

Punctuation	Effect on Avatar	Use For
Comma (,)	Brief natural pause	Lists, transitions, measured rhythm
Period (.)	Full stop — lets ideas land	Ending complete thoughts
Ellipsis (...)	Longer dramatic pause	Building anticipation before an important point
Paragraph break	Section-level pause	Separating distinct thoughts in one scene

✏️ Try It Now

Using the five-scene structure and the templates above, write your complete first video script right now. Target: 50–100 words per scene. Do NOT open Synthesia yet — finish the script first. When done, read every scene aloud. Fix anything that makes you stumble. Then move to Module 3.

Module 2 — Check Your Thinking

1. Why does copying text from a document into Synthesia almost never work well?

Synthesia has a word limit per scene that documents usually exceed Written language is structured for reading, not listening — it sounds robotic and dense when an AI delivers it at a fixed pace Documents use too many technical terms that Synthesia cannot pronounce

2. You are writing Scene 1 of your first video. Which opening is better and why?

"Welcome to this training on patient intake. In this video, we will cover the steps involved in processing a new patient form." — It sets clear expectations. "A patient is on hold. They've been transferred twice. They want to get booked — today. In the next four minutes, you'll learn the exact steps to make that happen every time." — It opens with a problem that earns attention.

3. What does Scene 4 (The Watch-Outs) contain?

A summary of everything covered in Scenes 1–3 The core step-by-step process in detail Common mistakes, exceptions, and what to do when things go sideways

Before You Open Synthesia — Pre-Flight Check

Confirm every item before creating a single scene:

☐	My 5-scene script is complete and approved
☐	I have read every scene aloud and fixed anything that sounds unnatural
☐	I know my video's title (format: [Topic] — [Title] — v1)
☐	I know which folder to save it in

If any of these are not checked, go back to Module 2. A video built on an unfinished script will always need to be rebuilt.

Step 1 — Create a New Video

Log in to Synthesia. Click New Video. Choose Blank — not a template. Before you do anything else, rename the video at the top of the editor. Use your naming convention: [Topic] — [Title] — v1. Save it into your folder.

💡 Pro Tip

The most common organizational mistake is building 10 videos before setting up a folder system. Set it up on Day 1. You will not regret it.

Step 2 — Configure Scene 1 (Your Configuration Scene)

Your first scene is your master template. Get these right before building more — every subsequent scene will be duplicated from this one.

Choose a layout. For instructional content, start with split-screen: avatar on one side, slide content on the other. This is the most versatile layout for training.
Enter your Scene 1 script. Paste your Hook scene (40–60 words) into the script text field. Just Scene 1 — not all five.
Choose an avatar. Filter by style: Professional, Casual, or Technical. Pick one appropriate for your topic. Upper-body avatars — showing the presenter waist up — work best for instructional content.
Choose a voice. Critical: do NOT just listen to the sample phrase. Paste your actual Scene 1 script into the voice preview field and listen. The voice that sounds great on "Welcome to today's training" may sound rushed on your actual content.
Preview Scene 1. Watch the full preview. Check: lip-sync accuracy, narration pace, and whether anything sounds off. Fix any issues before duplicating.

🏁 Quick Win

Once Scene 1 looks and sounds right, right-click it in the scene panel and select Duplicate. This gives you Scene 2 with your layout, avatar, and voice already configured. Change only the script. Repeat for all five scenes. This is the fastest way to build a consistent video.

Step 3 — Build Scenes 2 Through 5

For each remaining scene:

Duplicate the previous scene from the left panel
Replace the script with your next scene's content
Add a short on-screen text element — a headline (3–6 words), a numbered list of steps, or a key term. Not a transcript. A visual anchor.
Label the scene: right-click → rename. Use: Hook, Context, How-To, Watch-Outs, Landing
Preview the scene before moving to the next

📌 Note

On-screen text is not a transcript of what the avatar is saying. It is a visual anchor — short, scannable in 4 seconds, highlighting only the key point. If your slide has more text than your learner can read in 4 seconds, it has too much text.

Step 4 — Generate

Review all five scenes. Confirm each script is correct. Check on-screen text for spelling errors. Then click Generate in the top right.

Synthesia will process your video. Expect 5–15 minutes for a 5-scene video. You will receive a notification when it is ready. Do not sit and watch a progress bar — use the time to plan your next video or draft your second script.

Step 5 — Review and Fix

Watch the entire generated video before sharing it with anyone:

☐	Lip-sync looks accurate throughout all scenes
☐	Narration pace feels natural — not rushed, not sluggish
☐	On-screen text is fully visible — nothing cut off or overlapping the avatar
☐	No mispronounced words or robotic-sounding phrases
☐	Visual design is consistent across all five scenes
☐	No spelling or grammar errors in on-screen text

If you find a script error in one scene: open that scene → fix the text → click "Regenerate Scene." Only that scene re-renders (1–3 minutes). The rest of your video is untouched.

⚠️ Watch Out

Never share a first-generation video without watching it yourself first. A 2-minute personal review prevents the embarrassment of a learner catching your errors first.

Step 6 — Publish

Once your video passes the review checklist: go to your video in the library → click Share. Synthesia generates a hosted player link — a clean, professional video page anyone can watch in a browser.

For your first video, share the hosted link. That is the fastest path to done. Upload to your LMS, set up SCORM tracking, or build a formal course later. Get the video in front of learners today.

🏁 Quick Win

You just built your first Synthesia video. That is the hardest part — not because it is technically difficult, but because it is the point where most people hesitate too long. You didn't hesitate. Module 4 will make your next video look significantly more professional.

Module 3 — Check Your Thinking

1. What is the correct order of steps for building your first Synthesia video?

Open Synthesia → Choose avatar → Write script → Generate → Publish Finish and approve script → Create video → Configure Scene 1 → Duplicate for scenes 2–5 → Generate → Review → Publish Choose a template → Replace the text → Generate → Publish

2. Why do you preview your actual script text when choosing a voice — not just the default sample phrase?

The sample phrase is in a different language A voice that sounds great on a generic phrase may sound rushed or flat on your specific content — you need to hear your actual words Synthesia requires you to use the preview before you can select a voice

3. You catch a script error in Scene 3 of your generated 5-scene video. What is the most efficient fix?

Delete the video and rebuild from scratch Fix the error in Scene 3 and regenerate the full video Fix the script in Scene 3 only and click "Regenerate Scene" — only that scene re-renders

Why Design Matters More Than You Think

Learners make a quality judgment within the first 5 seconds of a video. That judgment is largely visual. A video with sloppy design — mismatched fonts, cluttered slides, no visible brand — signals "someone threw this together." A video with clean, consistent design signals "this organization takes its training seriously."

You do not need to be a designer. You need to follow five rules consistently.

First: Set Up Your Brand Kit (Do This Before Anything Else)

The Brand Kit stores your organization's visual identity and applies it to any video automatically. Set it up once — it works everywhere.

Logo

Upload as a PNG with transparent background. Transparent PNG = no white rectangle around your logo on any background. Size: at least 400px wide. If you only have a JPG, get the PNG from your marketing team before proceeding.

Brand Colors

Enter as hex codes (example: #1B2A4A). Include: primary color, secondary color, and a neutral for text. If you don't know your hex codes, check your brand style guide or ask marketing. Do not guess.

Fonts

Synthesia uses Google Fonts. Choose the closest available match to your brand fonts. Document the substitution so your whole team makes the same choice every time.

To apply: open any video → video settings → enable "Use Brand Kit." Every scene updates instantly. Enable this on every video from now on.

💡 Pro Tip

Screenshot your completed Brand Kit settings and share with your team. This means everyone configures their brand correctly without hunting through settings independently.

The Five Visual Rules

Rule 1 — One Idea Per Slide

Every slide should have one dominant element. One headline. One key statistic. One numbered list. One image. If two ideas are competing for attention on the same slide, split them into two scenes. Your learner cannot absorb two competing focal points simultaneously.

Rule 2 — Less Text Than You Think You Need

The slide is not a script. If your slides echo everything the avatar says word for word, your learner doesn't know where to focus — and they disengage from both channels. A slide with three short bullets readable in 4 seconds, then attention back to the avatar. That is the right rhythm.

Rule 3 — Consistent Layout Across the Video

Choose 2–3 layouts and use them consistently. Do not switch layouts every scene — it creates visual noise without purpose. Change layout only to signal a shift in content type: standard instruction uses Layout A, knowledge check uses Layout B. That is the full system.

Rule 4 — High Contrast Text — Always

Dark text on light background, or light text on dark background. Either works. What never works: medium-gray text on a light background, light blue on white, or any text that requires squinting. Your learners are watching on everything from a 27-inch monitor to a phone screen. When in doubt, increase the contrast.

Rule 5 — Your Logo, Every Video, Same Spot

Place your logo in the same corner of every video — top left or bottom right, small enough not to compete with content. Establish this in Scene 1 and carry it through duplication. Do not move it based on what looks good per scene. Consistency beats optimization.

The Five Most Common First-Video Mistakes

Mistake	What It Looks Like	The Fix
Too much slide text	Slides echo the full script; learners don't know where to focus	Remove 60% of the text. Keep only key terms, headlines, or step numbers.
No visual hierarchy	Everything on screen is the same size and weight	Make one element dominant. Make everything else smaller and lighter.
Inconsistent layouts	Different layout every scene, no clear pattern	Pick 2–3 layouts. Use them consistently. Change only with purpose.
No brand applied	Video looks like a Synthesia demo, not your organization	Set up Brand Kit before building. Apply to every video.
Logo with white box	Logo sits on a visible white rectangle over the background	Upload as transparent PNG. Get the PNG from marketing if needed.

✏️ Try It Now

Open the video you built in Module 3. Apply your Brand Kit. Then go through each scene and apply all five visual rules: reduce text, confirm hierarchy, check logo placement, confirm layout choices. Regenerate any scenes where you also changed the script.

Module 4 — Check Your Thinking

1. What file format should your logo be for Synthesia, and why?

JPG — it is the most widely supported format PNG with transparent background — so it sits cleanly over any background without a visible rectangle SVG — it scales without quality loss

2. Your slides in Module 3's video echo the full script text word for word. What should you do?

Leave it — learners benefit from seeing and hearing the same content simultaneously Remove about 60% of the slide text, keeping only key terms, headlines, or step numbers as visual anchors Remove all slide text — the avatar narration is sufficient

Step 1 — Add Captions Before You Publish Anything

Captions serve learners who are deaf or hard of hearing, learners watching without audio, learners for whom your video's language is a second language, and everyone whose comprehension improves with both channels active. Add them to every video.

Open your finished video in Synthesia
Find the Captions or Subtitles option in video settings or the sharing panel
Enable auto-generated captions
Review them in Synthesia's caption editor — AI is accurate but not perfect. Check proper nouns and technical terms.
Save the corrected captions
Export the SRT file from the download options — you will need this for LMS upload

📌 Note

An SRT file is a plain text file containing your caption text with timestamps. Every LMS and every major video platform accepts SRT files. Always export it alongside your MP4.

Step 2 — Choose How to Publish

Method	How It Works	Best When
Synthesia Hosted Link	Synthesia generates a shareable URL. Anyone with the link can watch in a browser.	Sharing for review, sending to a small group, embedding on a webpage
MP4 Download	Download the video file (up to 1080p). Upload anywhere.	Uploading to an LMS, sharing via drive, embedding in another platform
iFrame Embed	Copy an embed code from Synthesia. Paste into your LMS or webpage.	You want the video to stream from Synthesia inside your LMS without uploading a file

For your first video: use the Synthesia hosted link. Share it with your team. Get feedback. This is the fastest path to done. Upload to your LMS, set up SCORM tracking, and build formal course structure later. Do not let the perfect be the enemy of the published.

SCORM — The Short Version

You may have heard the word SCORM in relation to your LMS. Here is what you need to know right now:

What SCORM does: it lets course content communicate with your LMS — reporting whether a learner started, completed, and scored on content. Without SCORM, your LMS only knows a learner clicked on the content, not whether they watched or learned anything.

What Synthesia does: Synthesia exports MP4 video files and hosted player links — not SCORM packages natively. A Synthesia video by itself cannot report SCORM completion to your LMS.

Your Three Options

Option A — SCORM Wrapper: Download your Synthesia MP4 and wrap it in a SCORM shell using a free tool like iSpring Free. This turns your video into a trackable SCORM package. This is exactly how the companion Synthesia Masterclass SCORM course is built.
Option B — Separate Quiz Trigger: Host the video in Synthesia and use a short quiz on your LMS as the official completion event. The video is informational; the quiz generates the completion record.
Option C — LMS Basic Video Tracking: Many LMS platforms can track completion based on percentage watched when you upload an MP4 directly. Check your specific LMS documentation.

📌 Note

For most teams getting started, Option C (direct MP4 upload) or Option A (iSpring Free wrapper) is the right first step. Do not let SCORM complexity prevent you from publishing. Get the video published. Refine the tracking later.

Planning Your Second Video — Faster

Now that you have built one video, your second will take half the time. Your Brand Kit is configured, your avatar and voice are chosen, your folder structure exists, and you know the five-scene structure.

Open your first video. Right-click → Duplicate the entire project.
Rename the duplicate with the new video title.
Replace the script in each of the five scenes with your new content.
Update any scene-specific visuals (images, on-screen text).
Preview Scene 1. Generate. Review. Publish.

Your second video should take 60–90 minutes from script to published. Your fifth should take 45 minutes. Your tenth should take 30. The system gets faster every time you use it.

🎯 Real Talk

The single most important thing you can do after this course: build your second video this week. Not next week. This week. The gap between Video 1 and Video 2 is where momentum dies. Close it fast.

Your Production Checklist — Use This Every Time

☐	Script complete, read aloud, approved
☐	Video named: [Topic] — [Title] — v1
☐	Video saved in correct folder
☐	Brand Kit applied
☐	Avatar and voice previewed with actual script text in Scene 1
☐	All five scenes labeled: Hook, Context, How-To, Watch-Outs, Landing
☐	Generated video watched in full before publishing
☐	Captions enabled and reviewed
☐	SRT caption file exported
☐	Video published and link or file shared with intended audience

What Comes Next

You have completed the Synthesia Quick Start. You have built and published your first video, applied your brand, and have a checklist and production system you can use going forward.

Here is your next sequence:

This week: Build your second video using the duplicate-and-replace method.
Next week: Set up three scene templates in your Synthesia library — intro, content, knowledge check — so you stop configuring from scratch.
When ready: Take the Synthesia Masterclass to go deeper. Advanced scripting, visual design principles, translation, SCORM, team workflows, analytics, and a full Capstone Project.

Module 5 — Final Check

1. Synthesia natively exports which of the following?

SCORM 1.2 packages ready to upload to any LMS MP4 video files and shareable hosted player links xAPI (Tin Can) packages

2. What is the fastest way to build your second Synthesia video?

Start from a blank canvas and rebuild everything from scratch to ensure no errors carry over Duplicate your first video, rename it, and replace the script and visuals in each scene Use a Synthesia stock template from the library

3. Why should you add captions to every Synthesia video?

Captions are required by Synthesia's terms of service Captions serve learners who are deaf or hard of hearing, learners watching without audio, and learners where the video language is a second language — improving accessibility and comprehension for everyone Captions improve the video's generation quality in Synthesia

🎬

Quick Start Complete!

—

You have completed all 5 modules of the Synthesia Quick Start.

Your completion and score have been reported to your LMS.

Your next moves:
1. Build your second video this week
2. Set up three scene templates in your library
3. Take the Synthesia Masterclass when you're ready to go deeper

Module 1 — How Synthesia Actually Works

Learning Objectives

What Synthesia Is — And What It Is Not

What It Is Good For

What It Is Not Good For

Dashboard Orientation

The Scene — The Most Important Concept in Synthesia

How Long Is a Scene?

Module 1 — Check Your Thinking

Learning Objectives

The One Rule That Changes Everything

The Five Rules of Spoken English for Synthesia

The Five-Scene Structure

Script Templates — Fill In the Brackets

Controlling Pacing With Punctuation

Module 2 — Check Your Thinking

Before You Open Synthesia — Pre-Flight Check

Step 1 — Create a New Video

Step 2 — Configure Scene 1 (Your Configuration Scene)

Step 3 — Build Scenes 2 Through 5

Step 4 — Generate

Step 5 — Review and Fix

Step 6 — Publish

Module 3 — Check Your Thinking

Why Design Matters More Than You Think

First: Set Up Your Brand Kit (Do This Before Anything Else)

Logo

Brand Colors

Fonts

The Five Visual Rules

Rule 1 — One Idea Per Slide

Rule 2 — Less Text Than You Think You Need

Rule 3 — Consistent Layout Across the Video

Rule 4 — High Contrast Text — Always

Rule 5 — Your Logo, Every Video, Same Spot

The Five Most Common First-Video Mistakes

Module 4 — Check Your Thinking

Step 1 — Add Captions Before You Publish Anything

Step 2 — Choose How to Publish

SCORM — The Short Version

Your Three Options

Planning Your Second Video — Faster

Your Production Checklist — Use This Every Time

What Comes Next

Module 5 — Final Check

Quick Start Complete!