Q&A with KUDO Head of Innovation and AI Claudio Fantinuoli
Over the last eight months, Claudio Fantinuoli, Head of Innovation and AI at KUDO, has been working with his team to bring KUDO’s latest product, KUDO Interpreter Assist, to life. Claudio and team developed KUDO Interpreter Assist to aid interpreters in their preparation process. KUDO Interpreter Assist is a product with a line of features, the first creates high-quality, shareable, modifiable glossaries using Artificial Intelligence methods of Machine Learning and Natural Language Processing. We’ve asked Claudio to share some insights on the development of KUDO Interpreter Assist’s glossary builder and the other AI features on the horizon.
What sparked the idea for interpreter assist?
In a professional world that is – directly or indirectly – more and more influenced by technology, especially AI, we wanted to make use of the latest advancements in Natural Language Processing (NLP) to facilitate some aspects of the interpreting profession. The idea is to turn the potential of AI into a useful tool that can support interpreters in their daily work, without radically changing the way they operate.
What kind of interpreter research contributed to the development of this AI tool?
While designing the tool we combined and distilled the knowledge into two main areas of research:
- The latest advancements in the domain of NLP
- Computer-assisted interpreting
KUDO’s internal language department also influenced the design, proposing features and functionalities that are particularly useful in the context of online meetings.
What upcoming AI feature excites you the most?
I would have to say the upcoming Automatic Suggestion feature that will be released in the next few months. What fascinates me is the feature’s ability to show real-time information that may be useful to the interpreter, such as the terminology that a client desires for their meeting. We know that in some domains – for example in corporate meetings – clients require interpreters to use a specific industry jargon, even if alternatives may be completely fine and functional. Our tool will help meet this requirement. Interpreter Assist can be used or not depending on the circumstances, the interpreter’s attitude, the time they had to prepare a specific event, and so forth. I believe it will be a useful add-on to any interpreter’s toolbox.
Finally, in what other ways do you see AI facilitating multilingual communication in the future?
The application of AI in the language domain will increase the accessibility of information, both in the monolingual and in the multilingual space. This will foster inclusiveness, which is a noble goal for our modern societies. Think about the automatic captioning of meetings giving accessibility to people who are deaf or hard of hearing, the automatic translation of signed language into written text, or the translation of spoken language into another language. Notwithstanding the many limitations of AI, some of these applications are already there. In the foreseeable future they will not replace human interpreters, of course, but they will be good enough for all of the situations where currently no human professional is employed. This will offer some level of accessibility that is not yet available, and I think it will be a great achievement.
Interested in reading more about the future of AI and human interpreters? See Will AI Eventually Replace Human Interpreters? A Message from KUDO’s Chief Language Officer Ewandro Magalhaes