QUICK INFO BOX

Introduction

Video content is king, yet 73% of businesses can’t produce videos at scale. Corporate video production requires:

1. Studio booking ($500-2K per day)
2. Professional actors/presenters ($1K-5K per day)
3. Camera crew and equipment ($2K-5K per day)
4. Post-production editing ($1K-3K per project, 5-10 days)
5. Localization/translation ($500-1,500 per language)

Result: $10K-50K per corporate training video, 2-4 weeks production time, and astronomical costs for multilingual content. A Fortune 500 company creating 100+ training videos annually spends $1M-5M on video production alone. Updating content when products change requires reshooting—repeating entire expensive cycle.

Enter Synthesia, the AI video generation platform creating professional videos from text scripts in minutes—no cameras, actors, studios, or editing required. Founded in 2017 by Victor Riparbelli (CEO), Steffen Tjerrild (CTO), and AI researchers Lourdes Agapito and Matthias Niessner from University College London (UCL) and Technical University of Munich (TUM), Synthesia pioneered AI avatars—photorealistic digital humans that speak any script in 140+ languages with synchronized lip movements, natural gestures, and emotional expression.

As of February 2026, Synthesia operates at a $1.9 billion valuation with $156.6 million in funding from Accel, NVIDIA, Kleiner Perkins, GV (Google Ventures), FirstMark Capital, and MMC Ventures. The platform serves 50,000+ companies (February 2026) including Reuters, BBC, Zoom, Teleperformance, BSH Home Appliances, and enterprises across training, marketing, communications, and sales. Synthesia’s annual recurring revenue (ARR) exceeds $75 million (February 2026), making it the leading enterprise AI video platform.

With 350+ employees, 160+ diverse AI avatars (ethnicities, ages, accents, professional attire), and custom avatar creation (turning executives into AI versions), Synthesia has become essential infrastructure for corporate video production. The company’s Enterprise Studio platform allows teams to create, collaborate, manage, and distribute thousands of videos—replacing traditional video departments with AI-powered workflows.

What makes Synthesia revolutionary:

1. Text-to-video: Type script, select avatar/voice, generate professional video in 5-10 minutes—no filming required
2. 160+ AI avatars: Diverse presenters covering ethnicities, ages, professions—eliminating actor hiring
3. 140+ languages: Automatic translation and voicing—single script becomes 140 localized videos (vs. $500-1,500 per language traditionally)
4. Custom avatars: Recording 5-10 minutes of executive footage, creating AI version—CEO can “present” 1,000 videos without reshooting
5. Enterprise collaboration: Teams creating/editing videos together, brand guidelines enforced, analytics tracking engagement

The market opportunity spans $50+ billion video production market, $370+ billion corporate training industry, $600+ billion marketing/advertising, and $150+ billion internal communications. Every enterprise with employees needs training videos, product demos, sales enablement, executive communications—creating massive addressable market.

Synthesia competes with Runway ($4B valuation, text-to-video but consumer-focused), D-ID ($45M funding, talking head videos), Hour One ($26M funding, AI video), Elai.io ($6M funding), traditional video agencies ($10K-50K per video), and internal video teams (6-12 weeks per video). Synthesia differentiates through enterprise focus (SOC 2, SSO, brand controls), avatar diversity (160+ choices), collaboration features (teams working together), and language breadth (140+ languages vs. competitors’ 20-50).

The founding story reflects academic-to-commercial transition: UCL and TUM researchers pioneering computer vision techniques for photorealistic human synthesis saw opportunity to commercialize academic breakthroughs for enterprise video production. After demonstrating technology that allowed amateur actor to lip-sync as professional actor in any language, they founded Synthesia to democratize video creation.

This comprehensive article explores Synthesia’s journey from computer vision research lab to the $1.9 billion AI video platform transforming how 50,000+ enterprises create training, marketing, and communications content.

Founding Story & Background

The Corporate Video Problem

By 2016, video had become essential corporate communication medium—employees retain 95% of message when watching video vs. 10% when reading text (Forrester Research). Yet creating corporate videos at scale was prohibitively expensive:

Traditional video production workflow:

Phase 1: Pre-production (1-2 weeks)
Scripting, storyboarding, actor casting, location scouting, equipment rental. Cost: $3K-5K.

Phase 2: Production (1-3 days)
Studio booking, camera crew (3-5 people), lighting, sound, makeup, multiple takes, reshoots. Cost: $5K-15K.

Phase 3: Post-production (1-2 weeks)
Editing, color grading, graphics, music, voiceover, revisions. Cost: $5K-10K.

Total: $13K-30K per 3-5 minute video, 3-4 weeks timeline.

For enterprises creating 50-100 training videos annually (product launches, compliance, sales enablement), this meant $650K-3M budgets and dedicated video teams. Worse: localization—translating videos into 10-20 languages—required reshooting with native speakers or expensive dubbing ($500-1,500 per language), multiplying costs 10-20x.

When content updated (product features changed, compliance regulations updated), entire videos required reshooting—making video unsustainable for dynamic content.

Victor Riparbelli, entrepreneur with background in video production and technology, experienced these pain points while consulting with enterprises. Riparbelli saw pattern: companies needed 10x-100x more video content but couldn’t afford 10x-100x video budgets.

Meanwhile, at University College London (UCL) and Technical University of Munich (TUM), AI researchers Lourdes Agapito (professor, computer vision expert) and Matthias Niessner (professor, graphics/AI) were pioneering neural rendering—using deep learning to synthesize photorealistic humans. Their research enabled:

• Face reenactment: Transferring facial expressions from one person to another
• Lip synchronization: Matching lip movements to any audio track
• Style transfer: Modifying appearance, lighting, background

In 2016, they published research demonstrating Face2Face—real-time facial reenactment system capturing one person’s expressions and applying to another person’s face. The technology had disturbing deepfake potential but also legitimate use case: synthetic actors.

2017: Founding and Vision

In 2017, Riparbelli partnered with Agapito, Niessner, and engineer Steffen Tjerrild to found Synthesia (originally called “Visual Voice”) in London. The founding team combined:

• Victor Riparbelli (CEO): Business development, enterprise sales, video production expertise
• Steffen Tjerrild (CTO): Engineering, system architecture, scaling
• Lourdes Agapito (Chief Scientist): Computer vision, academic research, AI ethics
• Matthias Niessner (Chief Scientist): Graphics, neural rendering, deepfake detection (ironically, to prevent misuse)

The founding vision: Replace camera-based video production with AI-generated video—enabling anyone to create professional videos from text scripts.

Technical approach:

1. Record diverse actors (5-10 minutes of footage per actor)
2. Train neural networks on facial movements, voice, expressions
3. Create AI avatar that can speak any script with photorealistic lip-sync
4. Democratize video by eliminating cameras, actors, studios

The name “Synthesia” combined “synthesis” (creating content) and “synesthesia” (blending senses)—reflecting how AI synthesized visual and audio into coherent video.

2018-2019: Proof of Concept and Early Customers

From 2018-2019, Synthesia built first AI avatars:

Challenge 1: Photorealism
Early AI-generated faces looked uncanny—slightly off, triggering viewer discomfort. How to cross photorealism threshold?

Solution: High-quality recording (4K cameras, professional lighting), extensive training data (10-20 hours per avatar capturing every phoneme, expression), GAN-based rendering (refining output to eliminate artifacts).

Challenge 2: Lip Synchronization
Lip movements must perfectly match audio (phonemes). Misalignment creates obvious artificiality.

Solution: Phoneme-level synchronization—mapping audio to lip shapes frame-by-frame, training on 1,000+ hours of footage across languages.

Challenge 3: Natural Movement
Static avatars look frozen. Humans blink, shift weight, gesture naturally.

Solution: Motion models generating subtle movements—micro-expressions, eye movements, breathing—creating lifelike presence.

First customers were corporate training departments:

Use case: Global pharmaceutical company needing compliance training in 15 languages. Traditional cost: $250K (reshooting with native speakers). Synthesia cost: $15K (single script, 15 languages).

Result: 94% cost reduction, 2 weeks vs. 6 months production time, identical content across languages (no translation drift).

By 2019, Synthesia had 100+ enterprise customers and $2M ARR.

2020-2021: COVID Acceleration and Explosive Growth

The COVID-19 pandemic (2020) massively accelerated Synthesia’s growth. With lockdowns preventing traditional video shoots (no studios, no actors available), enterprises needed video alternatives for:

• Remote training: Onboarding employees virtually
• Internal communications: Executive updates without in-person meetings
• Sales enablement: Product demos without trade shows

Synthesia’s AI avatars required no physical presence—making it only viable solution.

Growth trajectory:

• 2019: 100 customers, $2M ARR
• 2020: 1,000 customers, $10M ARR (5x growth)
• 2021: 10,000 customers, $30M ARR (3x growth)

Synthesia launched custom avatars (2021)—allowing companies to create AI versions of executives, trainers, subject matter experts. Process:

1. Record executive for 10-20 minutes (reading script, multiple expressions)
2. Upload footage to Synthesia
3. AI trains custom avatar (2-3 days processing)
4. Executive can now “present” unlimited videos without ever filming again

Impact: CEOs creating quarterly updates in 20 languages, trainers building courses without repeated filming, sales teams personalizing demos at scale.

2022-2023: Enterprise Platform and Series C

From 2022-2023, Synthesia evolved from video generator to Enterprise Studio:

Features:

• Team collaboration: Multiple users editing videos simultaneously (like Google Docs)
• Template library: 60+ pre-built templates (product demos, training modules, sales pitches)
• Brand kits: Enforcing company colors, logos, fonts across all videos
• Analytics: Tracking video views, engagement, completion rates
• Integrations: Embedding into LMS (learning management systems), CRM, intranet

This transformed Synthesia from individual tool to enterprise infrastructure—replacing video departments with AI-powered workflows.

By 2023, Synthesia reached 30,000+ companies with $50M ARR—making it fastest-growing enterprise video platform.

Founders & Key Team

Victor Riparbelli (CEO) leads Synthesia with vision to democratize video creation. His enterprise sales expertise helped Synthesia win Fortune 500 customers early. Riparbelli is prominent thought leader on AI video, frequently speaking at conferences about responsible AI use.

Steffen Tjerrild (CTO) built Synthesia’s infrastructure serving 50,000+ companies generating millions of videos. His engineering enables real-time video generation (5-10 minutes from script to video).

Lourdes Agapito and Matthias Niessner (Chief Scientists) lead Synthesia’s AI research, advancing photorealism, language support, avatar diversity. Their academic backgrounds ensure Synthesia stays cutting-edge while addressing ethical concerns (deepfake prevention, consent models).

Funding & Investors

Seed (2018): $0.75 Million

• Lead Investor: Entrepreneur First (startup studio)
• Additional Investors: Angel investors (AI researchers, video professionals)
• Purpose: Building first AI avatars, early customer development

Series A (2019): $5.9 Million

• Lead Investor: Mark Cuban (Dallas Mavericks owner, Shark Tank investor)
• Additional Investors: LDV Capital, Seedcamp
• Valuation: ~$30M
• Purpose: Expanding avatar library, hiring engineering team, enterprise sales

Series B (2021): $50 Million

• Lead Investor: Kleiner Perkins
• Additional Investors: GV (Google Ventures), FirstMark Capital, MMC Ventures
• Valuation: $1 Billion (unicorn status)
• Purpose: Custom avatars, Enterprise Studio, international expansion

The Series B’s unicorn valuation reflected Synthesia’s 10,000+ customers, COVID-accelerated growth, and strategic importance in enterprise video market. GV’s (Google Ventures) investment signaled Google’s interest in AI video technology.

Series C (2023): $90 Million

• Lead Investor: Accel
• Additional Investors: NVIDIA, Kleiner Perkins, GV, FirstMark Capital
• Valuation: $1.9 Billion
• Purpose: Advanced AI features (emotional expression, gesture control), voice cloning, M&A

NVIDIA’s investment provided strategic value—GPU credits for training, technical collaboration, potential integration with NVIDIA’s Omniverse platform.

Series C Extension (2024): $10 Million

• Lead Investor: Accel
• Purpose: Geographic expansion (Asia-Pacific), language additions (reaching 140+ languages)

Total Funding Raised: $156.6 Million

Synthesia deployed capital across:

• AI research: Training advanced models, improving photorealism, expanding language support
• Infrastructure: GPU clusters (NVIDIA A100/H100), video rendering farms, serving 50K+ customers
• Avatar creation: Recording diverse actors (160+ avatars), processing pipelines
• Enterprise sales: Building SDR, AE, CSM teams for Fortune 500
• Compliance/security: SOC 2, GDPR, content moderation (preventing deepfake abuse)

Product & Technology Journey

A. AI Avatar Library (160+ Avatars)

Diverse representation across:

Ethnicities

• European, African, Asian, Middle Eastern, Latin American, Indigenous
• 50+ nationalities represented

Ages

• Young professionals (20s-30s)
• Mid-career (40s-50s)
• Senior/executive (60s+)

Professional Contexts

• Business formal (suits, corporate attire)
• Casual (smart casual, tech startup style)
• Industry-specific (healthcare scrubs, safety gear)

Accents

• Native accents for 140+ languages
• Regional variations (UK vs. US English, European vs. Latin American Spanish)

B. Custom Avatar Creation

Personalized AI versions:

Process:

1. Recording: 10-20 minutes of footage (reading provided script, multiple expressions, various head angles)
2. Upload: High-resolution video to Synthesia platform
3. Training: AI trains custom model (2-3 days, GPU-intensive)
4. Approval: Preview test videos, approve for production use
5. Deployment: Avatar available for unlimited video generation

Use cases:

• CEO communications: Quarterly updates, company announcements
• Subject matter experts: Technical trainers, product specialists
• Sales teams: Account executives personalizing demos at scale

Pricing: $1,000-3,000 per custom avatar (one-time), unlimited usage

C. Video Creation Workflow

Text-to-video in 5 steps:

Step 1: Script
Write or paste script (100-5,000 words), AI estimates video length

Step 2: Avatar Selection
Choose from 160+ avatars or custom avatar, select language/accent

Step 3: Voice Selection
Choose AI voice (natural, authoritative, friendly, etc.), adjust speed/pitch

Step 4: Visual Design

• Templates: 60+ pre-built layouts (corporate training, product demo, news broadcast)
• Backgrounds: Solid colors, gradients, uploaded images, video backgrounds
• Graphics: Text overlays, logos, charts, animations
• Music: Royalty-free background music library

Step 5: Generate
Click “Generate Video”, AI renders in 5-10 minutes (depending on length)

Output: MP4 video (1080p or 4K), downloadable or shareable via link

D. Multi-language Support (140+ Languages)

Automatic translation and voicing:

Process:

1. Write script in source language (e.g., English)
2. Select target languages (e.g., Spanish, French, German, Japanese, Arabic)
3. AI translates script (preserving meaning, cultural context)
4. Avatar speaks in target language with native accent and perfect lip-sync
5. Generate 10 videos in 10 languages simultaneously

Languages supported: All major languages (English, Spanish, French, German, Chinese, Japanese, Arabic, Hindi, etc.) plus 100+ regional languages (Swahili, Tagalog, Vietnamese, Thai, Hebrew, etc.)

Cost savings: Traditional localization costs $500-1,500 per language. Synthesia: $0 marginal cost.

E. Enterprise Studio Features

Team collaboration:

• Multi-user editing: Teams editing videos together (Google Docs-style)
• Commenting: Feedback threads on specific frames
• Approval workflows: Routing videos through stakeholders for sign-off
• Version control: Tracking changes, reverting to previous versions

Brand management:

• Brand kits: Company colors, fonts, logos enforced automatically
• Template libraries: Custom templates for recurring video types
• Asset management: Centralized storage for graphics, music, footage

Analytics:

• View tracking: Who watched, how long, completion rates
• Engagement metrics: Rewinds, pauses, drop-off points
• A/B testing: Comparing different video versions for effectiveness

Integrations:

• LMS: Canvas, Moodle, Cornerstone, SAP SuccessFactors
• CRM: Salesforce, HubSpot (embedding videos in email campaigns)
• Communication platforms: Slack, Microsoft Teams, intranet portals

F. AI Technology Stack

Computer Vision:

• GANs (Generative Adversarial Networks): Generating photorealistic faces
• Neural rendering: Synthesizing lighting, shadows, textures
• Facial landmark detection: Tracking 68+ facial points for accurate animation

Speech Synthesis:

• Text-to-speech (TTS): Converting text to natural-sounding speech (140+ languages)
• Voice cloning: Creating AI voice from 10-20 minutes of audio samples (custom avatars)
• Prosody modeling: Natural intonation, pauses, emphasis

Lip Synchronization:

• Phoneme mapping: Matching audio to lip shapes (visemes)
• Temporal coherence: Smooth transitions between frames
• Co-articulation: Accounting for how adjacent sounds affect lip movements

Infrastructure:

• GPU clusters: NVIDIA A100/H100 GPUs for training and inference
• Video encoding: Optimized MP4 encoding (H.264/H.265)
• CDN: Global content delivery for fast video loading

G. Ethical AI and Safety

Preventing deepfake abuse:

• Consent models: All avatars recorded with explicit consent, legal agreements
• Watermarking: Subtle watermarks identifying Synthesia-generated videos
• Content moderation: Blocking harmful content (hate speech, misinformation, impersonation)
• API restrictions: Preventing automated misuse, rate limiting
• Transparency: “Generated by AI” disclosures on videos

Governance:

• Ethics board: Overseeing responsible AI use
• Partnerships: Collaborating with MIT, Stanford on deepfake detection
• Industry leadership: Founding member of Content Authenticity Initiative (CAI)

Business Model & Revenue

Revenue Streams (February 2026)

Pricing Tiers:

• Personal ($30/month): 10 videos/month, standard avatars, 720p
• Business ($90/month): 40 videos/month, premium avatars, 1080p, custom templates
• Enterprise (custom): Unlimited videos, custom avatars, 4K, brand kits, SSO, dedicated support

Subscription Metrics (Estimated):

• 50,000+ companies
• Average $1,500 annual contract (weighted by tier)
• Annual revenue: $75M ARR

Customer Segmentation

1. Corporate training (40%): L&D departments creating employee training
2. Marketing (25%): Product demos, explainer videos, social media
3. Internal communications (20%): Executive updates, company announcements
4. Sales enablement (15%): Personalized sales videos, prospecting

Unit Economics

• CAC: $500-1,500 (inside sales, inbound marketing, product-led trials)
• LTV: $15,000+ (multi-year contracts, expanding seats, custom avatars)
• Gross Margin: 75%+ (GPU costs declining, economies of scale)
• Payback Period: 12-18 months
• Churn: 15% annually (high switching costs, integrated into workflows)

Total ARR: $75+ Million (February 2026), growing 50%+ YoY

Competitive Landscape

Runway ($4B valuation): Text-to-video but consumer/creator-focused, not enterprise
D-ID ($45M funding): Talking head videos, narrower feature set
Hour One ($26M funding): AI video platform, smaller scale
Elai.io ($6M funding): Budget AI video tool
Traditional video agencies: $10K-50K per video, 3-6 weeks
Internal video teams: 4-12 weeks per video, limited scale
PowerPoint/Keynote: Static presentations, no video

Synthesia Differentiation:

1. Enterprise focus: SOC 2, SSO, brand controls, enterprise SLAs
2. 160+ avatars: Largest diversity (competitors have 20-50)
3. 140+ languages: Broadest language support (competitors have 20-50)
4. Custom avatars: Turning executives into AI versions (unique to Synthesia)
5. Collaboration: Teams working together (not just individual creation)

Impact & Success Stories

Corporate Training

Teleperformance (customer service BPO, 420K employees): Using Synthesia to create training in 35 languages. Cost savings: $2M+ annually vs. traditional video. Production time: 85% reduction.

Media

Reuters (news agency): Using AI avatars for news updates in 12 languages. Publishing 3x more multilingual content with same team size.

Technology

Zoom (video communications): Using Synthesia for product demos, sales enablement. Sales cycle: -30% (video demos accelerating deals).

Future Outlook

Product Roadmap

Emotional intelligence: Avatars expressing emotions (excitement, empathy, concern) based on script sentiment
Full-body avatars: Moving beyond talking heads to full-body presenters with gestures, walking, interaction
Real-time generation: Live avatar streaming (virtual presenters in webinars, conferences)
Interactivity: Viewers asking questions, avatar responding (AI-powered Q&A)

IPO Timeline

With $75M ARR, 50%+ growth, 50,000+ companies, and $1.9B valuation, Synthesia positioned for IPO in 2027-2028 as AI video becomes standard in enterprise communications.

FAQs

What is Synthesia?

Synthesia is AI video generation platform creating professional videos from text scripts using 160+ AI avatars speaking 140+ languages, with custom avatar creation for enterprises.

How much does Synthesia cost?

Personal ($30/month, 10 videos), Business ($90/month, 40 videos), Enterprise (custom pricing, unlimited videos, custom avatars). Custom avatars: $1K-3K one-time fee.

What is Synthesia’s valuation?

$1.9 billion (February 2026) following a $90M Series C led by Accel with NVIDIA, Kleiner Perkins, GV participation.

How many companies use Synthesia?

50,000+ companies including Reuters, BBC, Zoom, Teleperformance, BSH Home Appliances, and Fortune 500 enterprises.

Who founded Synthesia?

Victor Riparbelli (CEO), Steffen Tjerrild (CTO), and AI researchers Lourdes Agapito (UCL) and Matthias Niessner (TUM), founded 2017 in London.

Conclusion

Synthesia has democratized professional video production, reducing costs from $10K-50K to $30-90/month and timelines from 3-6 weeks to 5-10 minutes. With a $1.9 billion valuation, $75M+ ARR, and 50,000+ companies worldwide, Synthesia has proven that AI-generated video is not experimental technology but enterprise-grade infrastructure—powering training, marketing, sales, and communications at global scale.

As AI models improve (photorealism approaching indistinguishable from real humans, emotional intelligence enabling nuanced expression), Synthesia’s avatars will become increasingly sophisticated. The company’s enterprise focus, ethical AI leadership, and comprehensive platform (collaboration, analytics, integrations) position it as video production standard for Fortune 500. With 50%+ growth, NVIDIA partnership, and Accel backing, Synthesia is positioned as compelling IPO candidate within 24-36 months, potentially achieving $5B+ public market valuation as every enterprise adopts AI video for scaled communications.

Related Article:

• https://eboona.com/ai-unicorn/6sense/
• https://eboona.com/ai-unicorn/abnormal-security/
• https://eboona.com/ai-unicorn/abridge/
• https://eboona.com/ai-unicorn/adept-ai/
• https://eboona.com/ai-unicorn/anduril-industries/
• https://eboona.com/ai-unicorn/anthropic/
• https://eboona.com/ai-unicorn/anysphere/
• https://eboona.com/ai-unicorn/applied-intuition/
• https://eboona.com/ai-unicorn/attentive/
• https://eboona.com/ai-unicorn/automation-anywhere/
• https://eboona.com/ai-unicorn/biosplice/
• https://eboona.com/ai-unicorn/black-forest-labs/
• https://eboona.com/ai-unicorn/brex/
• https://eboona.com/ai-unicorn/bytedance/
• https://eboona.com/ai-unicorn/canva/
• https://eboona.com/ai-unicorn/celonis/
• https://eboona.com/ai-unicorn/cerebras-systems/