What We Do
Document search
Avyana’s AI can digest multiple documents, understand questions posed in natural language and present the user with the document parts answering the question.
Q&A on technical manuals
Avyana’s AI can digest technical manuals, integrate them in its pre-existing model of the world and answer questions on them.
Teaching assistant
Avyana’s AI can assist teachers in the task of correcting homework: from a picture of the homework it will directly identify the errors and present a grade suggestion.
Avatars with Avyana's AI
A zero-code Artificial Intelligence that can learn and create an internal model of the world from a body of knowledge.
It will be the engine of Intelligent Avatars for:
- Industrial Metaverse
- Commercial Metaverse
- Health care sector
- Education and information
- Environment and climate change
Emotional connection
Bridge the uncanny valley by making the avatars aware of the emotions of the users, and responsive to them.
Natural language processing
Push the processing of natural language beyond the limits of text manipulation and towards the update and synchronization of a model of the world.
Image segmentation
Segment an image into components, each containing a single object, even if the AI has never been trained to recognize these particular objects.
Object Acquisition
Follow an object through all the frames of a video, through translations, rotations, and concealments.