research-background

Research and Development

Get in touch to become a research partner

Research and Development

Standing at the cutting edge of artificial intelligence, we at Oxolo have a dedicated research and development team whose goal is to push the boundaries of the current state-of-the-art in natural language processing, computer vision and speech recognition and synthesis.

Thus, it is no surprise to find us collaborating with universities and research institutes or at renowned international conferences such as NeurIPS, ICLR, ICML, CVPR and the like where we submit and review papers. Beyond continuously improving the quality and speed at which the interaction is synthesized, we research and develop algorithms to protect our system from harmful biases in the data to tailor AI-driven avatars that go through an extensive quality assurance procedure to reliably serve the users in a fail-safe and ethical way.

As almost all data on the internet is biased in multiple ways towards discrimination against gender, race, sexual preference, religion etc., we reach out to every righteous mind in AI to join us on our mission to make AI safe and fair for everyone.

Technology overview

Basic user audio input is processed and classified by our natural language processing engine, analyzed by our AI and combined with our computer vision machinery to deliver a fully rendered video chat with precise lip coordination and context-modulated speech.

User Speech Input

user-speech

Natural Language Processing

proccesser
computer-vision

Computer-Vision Machinery

generalized

Generalized

text-to-speech

Text-to-speech

specialized

Specialized

video-chat

Video Chat Output