Part of what makes KUDO AI unique is the Patented process behind our AI engine. We have an entire team in-house dedicated to improving the quality and user experience of this solution every day, and most recently, they’ve achieved this by integrating Large Language Models (LLM) in a global first for real-time speech translation technology.
Innovation doesn’t come without its challenges. As Claudio Fantinuoli, KUDO’s CTO, explains: “Controlling the behavior of the LLM isn’t easy, especially when our objective is to keep latency to a minimum. The huge advantage of using LLM, however, is that it allows us to translate using context (what a person has said in the original language) which in turn improves the quality of translation (coherence and accuracy). Our R&D focus on LLM puts us in a great position to improve translation quality further in the future, which is always our main goal at KUDO.”
Last, but certainly not least. Until now, AI speech translation has globally been limited to one-to-many use cases like live events, conferences, trainings, and internal company announcements. This is principally due to the increased latency incurred through AI translation compared with human interpretation, which would make for a poor user experience if you tried to hold a multi-way conversation using an AI speech translator.
This month, KUDO AI will become available as a fully conversational translation solution as well, through a specialized form of real-time subtitles that can be chosen upon creating a meeting or event. Users will have the option to choose either a webinar-style, one-to-many setting when they create their meeting – in which case translated audio and captions will be activated – or a conversational setting in which case live subtitles will be activated.
Without the added translation of audio, this format reduces latency (the time it takes for a translation to be rendered) to the absolute minimum while prioritizing accuracy and fluency. This means that back-and-forth conversations in multiple languages can be followed in the language of the user’s choice in as close to real-time as possible, expanding the use cases for KUDO AI to popular requests like board meetings, interviews, focus groups, Q&A sessions, and interactive conversations.
About KUDO
KUDO is the world leader in providing real-time multilingual solutions that enable people to communicate effortlessly in any language—on any platform. Their network of 12,000 professional language interpreters, combined with their ground-breaking Speech Translator, empower organizations of all sizes to collaborate more efficiently, with greater inclusivity, and on an international scale. KUDO Inc. is a New-York based technology start-up founded and managed by language and conferencing industry insiders seeking to create a world in which everyone has the power to understand and be understood in their own language. More info at www.kudoway.com.
Contact
Chloë Walford-Delahaye
chloe@kudoway.com