You type a script. You pick an AI avatar. Synthesia generates a polished video of that avatar delivering your content -- no camera, no microphone, no editing timeline.
That's the pitch. And after working with Synthesia across a range of use cases -- corporate training videos, product walkthroughs, multilingual explainers -- I can tell you it mostly lives up to it. There are tradeoffs worth understanding before you commit to a subscription, but the core product works.
This guide covers everything you need to get started: what Synthesia actually is, how to create your first video, how to navigate the interface, and what the pricing tiers get you.
For a deeper evaluation of how Synthesia stacks up as a product, see our full Synthesia review. For how it compares against alternatives, check the best AI video generators roundup.
What Synthesia Is
Synthesia is an AI video platform built around one core feature: AI avatars. You provide a script, choose from a library of realistic digital human presenters, and Synthesia renders a video of that avatar delivering your content.
The avatars are diverse -- different ethnicities, ages, presentation styles, formal and casual settings. The lip sync technology is the part that makes or breaks these tools, and Synthesia's is genuinely good. Not indistinguishable from a real recording, but well past the uncanny valley for professional/business video purposes.
What Synthesia is notably good at:
- Corporate training content -- consistent presenter, consistent format, easy to update
- Product demos and onboarding -- swap scripts when features change, no re-recording
- Multilingual content -- generate the same video in multiple languages without hiring talent for each
- Internal communications -- executive messages, HR announcements, policy updates at scale
What it's not ideal for: creative/cinematic content, entertainment, anything requiring genuine human expression or improvisation. It's a production efficiency tool, not a creative AI film generator.
The closest competitor is HeyGen, which offers similar avatar video capabilities with a stronger translation feature. Synthesia tends to win on avatar polish and enterprise feature depth; HeyGen wins on translation quality and video dubbing.
Synthesia Pricing
Before diving into the how-to, understanding the pricing structure matters because it affects how you'd use the product.
Starter plan ($29/month): 10 video credits per month. Each credit generates roughly one minute of video. The starter plan includes access to a limited avatar library and basic template options. No custom avatar creation.
Creator plan ($89/month): Unlimited video generation (within fair use limits), full avatar library access, custom branded avatars using your own footage, priority rendering, and advanced video editing within the platform. This is the tier most serious users land on.
Enterprise: Custom pricing, custom avatars, SSO, team management, compliance features. For teams producing significant video volume.
The free tier has been removed from Synthesia's current offering -- you can request a demo, but there's no true free plan as of early 2026. That's a meaningful difference from HeyGen, which still offers a free-credit free tier.
Setting Up Your Account
Go to Synthesia and create an account. You can sign up with email or SSO via Google. The onboarding flow walks you through account verification quickly.
Once you're in, you land on the Synthesia dashboard -- a clean interface with your recent projects, a "Create Video" button, and navigation to your templates library, avatar browser, and account settings.
Before your first video: take five minutes to explore the avatar library. Filter by style (professional, casual, formal), gender, and ethnicity. Not every avatar is equal in quality -- some have noticeably more natural expression and movement than others. Note two or three you'd want to use before you start your first project.
Creating Your First Video
1. Start a New Project
Click "Create Video" from the dashboard. Synthesia prompts you to choose a starting point:
- Blank video -- Build from scratch
- Template -- Start with a pre-designed layout with placeholder elements
- AI-generated script -- Paste a topic and let Synthesia draft a script before you review and edit
For your first video, I'd recommend starting with a template. Synthesia's templates are professionally designed and handle a lot of the layout work -- you just swap in your content. It's a faster path to seeing what a finished Synthesia video actually looks like.
2. Write Your Script
The script editor is where most of your work happens. Each scene in a Synthesia video has its own script block, corresponding to one chunk of what the avatar says.
A few things that matter a lot here:
Write for listening, not reading. If you're adapting content from a document or slide deck, don't paste the text directly. Document writing and spoken language are different. Long sentences that work fine on paper become a stumbling block for avatars. Short sentences, verbal transitions, natural breaks -- write the way you'd actually say it.
Use punctuation to control pacing. Commas create brief pauses. Periods create longer pauses. If you want a deliberate pause mid-sentence, an ellipsis (...) works. Synthesia's TTS engine uses punctuation for rhythm, not just grammar. A script with no pausing feels rushed. One with deliberate punctuation breathes.
Watch out for acronyms and brand names. "API" might be pronounced "app-ee" by the TTS engine if it doesn't know the word. Technical terminology, product names, and unusual proper nouns often need phonetic spelling in the script. Write "A-P-I" or "ay-pee-eye" if the pronunciation matters.
Target 130-150 words per minute for natural avatar delivery. A 200-word script produces roughly a 90-second video. Scale up or down accordingly.
3. Choose Your Avatar
The avatar selector is in the right panel of the editor. You can filter the library by:
- Gender
- Ethnicity
- Style (professional, casual, business, etc.)
- Camera angle (upper body, full body, close-up)
Click any avatar to see a preview clip. Watch for natural head movement, eye blinks, and expression variation -- these make avatars feel less robotic. The "premium" avatars, usually labeled as such, generally have better underlying motion quality.
Match the avatar to your content context. A formal business presentation works with a professional-attired avatar in a neutral background setting. An informal tutorial might work better with a casual presenter. The disconnect between avatar style and content tone is more noticeable than people expect.
Custom avatars -- using your own recorded likeness or someone from your team -- are available on Creator and Enterprise plans. Synthesia's custom avatar process requires recording a specific consent video following their guidelines. The result is a digital version of the real person, which some organizations find more credible for internal communications.
4. Select a Voice
Voice selection sits alongside avatar selection. Synthesia offers a large voice library organized by language, accent, and gender. Highlights:
Preview voices with your actual script text. The default preview uses a sample sentence that's been chosen to make the voice sound good. Your actual script -- especially if it contains unusual terms -- may sound different. Type a sentence from your real script into the preview before committing.
Match voice accent to content audience. If you're producing content for a UK audience, a British accent reads as more natural. If your avatar has a clearly South Asian presentation, an Indian English accent can feel more coherent. These are subtle signals, but they accumulate.
Voice quality varies. Synthesia's newer voices (often labeled "Neural" or with quality indicators) are noticeably better than older entries in the library. If a voice sounds mechanical even in preview, skip it.
For multilingual content: Synthesia supports video generation in 140+ languages. You can create separate scenes in different languages within the same project, or duplicate a project and swap the script and voice for each language version.
5. Design the Scene
Each scene has layout options beyond just avatar and script:
Background: Upload a branded image, use Synthesia's built-in background library, or set a solid color. Branded backgrounds (a tasteful office image, a background with your logo color) look more professional than virtual backgrounds for most business use cases.
Screen recordings and images: You can layer content behind or alongside the avatar -- product screenshots, charts, slides. This is useful for product demos where you want to show the interface while the avatar narrates.
Text overlays: Lower-thirds (name + title beneath the avatar), section titles, callout text. Keep text sparse. Too much text on screen with a talking presenter creates visual competition.
Templates handle most of this automatically. If you started from a template, the layout is already set up. You're mostly swapping content into existing placeholder elements.
6. Build Your Full Video
A Synthesia video is a sequence of scenes. Each scene has its own script, avatar position, background, and overlaid content.
Structure your video logically:
- Opening: brief intro, establish the topic
- Body: one main point per scene, ideally
- Close: summary, CTA, or what to do next
Keep the overall video focused. Synthesia works best for videos under 10 minutes. For longer content (full courses, multi-topic training), break it into individual shorter videos organized in a playlist or collection.
7. Preview and Adjust
Before rendering, use the scene-by-scene preview. This plays a rough preview of each scene -- not final render quality, but enough to catch obvious problems like:
- Scenes that are too long or too short
- Script that sounds awkward when previewed
- Visual elements that overlap or conflict
Fix issues at the preview stage rather than after rendering. Rendering a full video and realizing the script has a pacing problem halfway through costs you time.
8. Render Your Video
Hit "Generate" (or "Render," depending on the interface version). Synthesia renders your video in the cloud. Processing time varies:
- Under 3 minutes of video: typically 5-10 minutes to render
- 5-10 minute videos: can take 20-30 minutes
- Priority rendering (Creator plan): noticeably faster
You'll get an email when the video is complete. You don't need to leave the tab open.
When it's done: watch the full video before downloading. Check lip sync at sentence starts and ends, catch any mispronounced words, verify the scene flow makes sense end-to-end.
9. Export and Distribute
Synthesia exports to MP4 at 1080p by default. Options vary by plan:
- Download directly: Good for distribution to platforms where you upload your own files (YouTube, LinkedIn, internal LMS, etc.)
- Shareable link: Synthesia generates a link to a hosted version of your video -- useful for sending directly without needing to download and re-upload
- Embed code: For embedding the video in a web page or LMS
For most web distribution: 1080p MP4 is fine. If you're distributing to large screens or projectors, check whether your plan offers higher-resolution export.
Practical Tips for Better Videos
After creating a lot of Synthesia videos across different use cases, these are the things that consistently matter:
Keep scripts under 150 words per scene. Longer scripts don't fail, but they stretch individual scenes and the pacing starts to feel like a lecture instead of a presentation.
Use the same avatar consistently within a series. Consistency creates brand identity. If you're creating a training course with 10 modules, the same presenter across all 10 feels more professional than 10 different avatars.
Don't skip the preview step. It's easy to rush through and hit Render. The preview takes 30 seconds per scene and regularly reveals problems you'd miss in the editor.
Update existing videos efficiently. When a product feature changes and you need to update a demo video: you don't need to recreate it from scratch. Open the project, edit the script for the affected scenes, regenerate. This is one of Synthesia's most practical advantages over traditional recording.
Script reviews before recording. Have someone read the script aloud before you generate. They'll find awkward phrasing faster than you will because you're too close to the content. Thirty seconds of reading → less wasted render time.
Is Synthesia Worth It?
For the right use cases, clearly yes.
If you're producing corporate training content, onboarding videos, or multilingual explainers at any significant volume, Synthesia's ability to generate professional-quality presenter video without recording equipment changes the economics of video production. A video that would have required a studio booking, a presenter, and an editor now takes a couple of hours of script and setup work.
The $29 starter plan is reasonable for individual creators or small teams trying the product. If you hit the 10-credit ceiling quickly, the jump to Creator ($89/month) is the obvious upgrade.
The absence of a free tier is the biggest friction point for new users evaluating whether to commit. If you want to try Synthesia before paying, requesting a demo from their sales team is the path -- or using a trial if one's currently offered.
For alternatives: HeyGen offers a free credit tier if you want to explore AI avatar video at zero cost before committing. Our AI video generators roundup compares both in detail.
Ready to try it? Visit Synthesia to start your account and explore the platform.
Disclosure: TechSifted does not currently have an affiliate relationship with Synthesia. The link above is a direct, non-affiliate link -- we earn nothing if you sign up. Synthesia is on our affiliate application list for March 19, 2026. If we establish a relationship, this disclosure will be updated. This guide reflects our honest evaluation regardless of monetization.
Top comments (0)