Synthesia is a London-based software company that develops artificial intelligence (AI)-powered video generation technologies. Its platform enables users to convert text into video content without the need for cameras, microphones, or studio equipment. Founded in 2017, Synthesia provides photorealistic avatars, multilingual voice-over systems, and video editing tools, catering especially to the fields of education, marketing, customer service, and corporate communications.

Founding and Founders

Synthesia was established in 2017 by Prof. Matthias Niessner (Technical University of Munich) and Prof. Lourdes Agapito (University College London). The founding team commercialized their academic expertise in areas such as computer vision, 3D modeling, and neural video synthesis.

Core Technologies

At the heart of Synthesia’s platform is its “text-to-video” (TTV) generation system. This technology transforms written scripts into visual narratives featuring photorealistic human avatars. It employs neural network-based video synthesis that learns facial expressions, vocal tone, and body language to produce realistic video output. The system aims to avoid the "uncanny valley" effect often associated with digital humans.

AI Avatars

Synthesia offers over 230 ready-made AI avatars representing a wide range of genders, ages, and ethnicities. Users can also create custom avatars based on their own voice and appearance, which are capable of speaking in over 30 languages. “Studio avatars,” produced with advanced recording techniques, provide higher resolution and more natural motion quality.

AI Voice Technologies

The platform allows text to be converted into natural-sounding, multilingual voice-overs using hundreds of AI voice options. A voice cloning feature enables users to generate digital replicas of their own voices. Synthesia supports over 140 languages for automatic translation and dubbing. Its beta “AI Dubbing” feature translates existing videos while preserving the original speaker’s voice and speaking style.

Video Editing and Production Tools

Synthesia enables content creation without the need for complex editing software. The platform includes over 300 pre-designed templates, a drag-and-drop editor, a media library (with images, videos, music, and GIFs), a screen recorder, and text-to-video tools. The AI Video Assistant can automatically convert presentation files, web pages, or documents into video content.

Enterprise Features

For corporate users, Synthesia provides unlimited video production, real-time collaborative editing, brand identity tools, and data analytics. The platform complies with SOC 2 and GDPR standards, ensuring secure content management and ethical AI usage.

Synthesia integrates with numerous third-party tools including PowerPoint, WordPress, Shopify, Moodle, Vimeo, and HubSpot. It is optimized for compatibility with learning management systems and sales enablement tools.

Research and Open Source Contributions

The company contributes to academic research in areas such as natural facial expression generation, speech synchronization, and photorealism. Synthesia has released the HumanRF dataset (a four-dimensional human capture dataset) and ActorsHQ, a multi-camera human motion database, to support the development of 3D neural rendering techniques and cinematic-level digital content creation.

Ethical Principles and Content Moderation

Synthesia places AI ethics at the core of its corporate policy. It guarantees that no individual representation is created without consent. All content is reviewed before publication through both AI-based and human moderation processes. Synthesia is a member of the Content Authenticity Initiative led by Adobe.

By automating video production through text-to-video technology, Synthesia reduces the need for traditional video production tools. The platform is widely used in education, sales, customer service, and content marketing, offering scalable, multilingual content creation via AI-powered avatars and voice synthesis.

Future Outlook

Synthesia’s development strategy aims to broaden access to AI-based video production and digitize media creation workflows. The research team is working toward systems capable of generating not only digital avatars but also fully synthetic scenes—reproducing not just individuals but also their environments and contexts.

The goal is to allow users to generate accurate, multilingual, interactive, and context-aware videos solely from text input. Strategic priorities also include energy efficiency, ethical production standards, and digital content security. Synthesia aims to provide AI-based video synthesis solutions capable of fully replacing physical camera setups and continues its work in collaboration with academic institutions, media organizations, and regulatory bodies.