NC, Publishes Papers for 4 Consecutive Years at Prestigious Global Voice AI Conference

"Voice AI Technology Recognized as World-Class"

NCSoft announced on the 18th that it has published papers at the globally prestigious speech AI conference 'INTERSPEECH' for four consecutive years.

Celebrating its 24th anniversary this year, INTERSPEECH is one of the largest conferences in the field of speech AI worldwide. It brings together experts in natural language processing (NLP), speech recognition AI, as well as linguists and phoneticians from around the globe.

NCSoft's AI Center's 'Speech AI Lab' has been sharing its research results at world-renowned AI conferences. It has published papers at INTERSPEECH for four consecutive years, with three papers accepted this year as well.

The topics of this year's published papers include ▲user-customized wake word recognition technology ▲development of a new multimodal emotion recognition technology that significantly improves existing emotion recognition methods ▲a high-speed, lightweight wake word recognition model that can run on web browsers without server communication. These papers are meaningful as they provide foundational technologies for achieving natural interaction between humans and AI.

Kim Hee-man, head of NCSoft's Speech AI Lab, said, "This is an example that shows NCSoft's speech AI technology is already world-class. Moving forward, we will conduct research on auditory intelligence that encompasses not only speech technology but all sounds, contributing to the realization of true digital humans."

This content was produced with the assistance of AI translation services.