#ncsoftcom
**NCsoft secures digital human competitiveness… Singing AI development**
NCsoft is making achievements in the field of artificial intelligence (AI) speech synthesis. It is evaluated that it is securing technological competitiveness through steady R&D and technology development. It seems to be securing digital human technology capabilities in the future through sound and visual investments.
According to the game industry on the 30th, NCsoft recently established a music AI team in the speech AI lab in the AI â??â??center. It is a team of researchers who develop AI models and musicians who build and analyze music data. The goal is to achieve high-quality results by working together based on their expertise.
The music AI team’s initial research focused on the ‘singing voice’, a voice that naturally sings. However, he expanded the scope to the whole of music with various elements, such as creating melodies and creating harmony. Currently, we are conducting research in several fields at the same time so that AI can be applied to music. The NCsoft Music AI team consisted of practical music experts. Through musical expertise, the quality of data was improved and the AI â??â??song made it possible to pursue naturalness, such as feeling emotional lines.
NCsoft is sharing the results of AI speech synthesis-related research with various academic societies. This month, three papers were presented at Interspeech 2022, the world’s most prestigious speech AI conference where global AI companies share the latest technologies. The three papers published this time are based on adversarial multi-task learning, a study on separate modeling of tone and pitch expression, a study on improving speech quality using pre-learned neural vocoders, research, etc. Through this research result, NCsoft was evaluated to have realized the quality and naturalness of the voice synthesis and song synthesis system at the highest level in the industry.
NCsoft is accelerating the preparation of digital humans by strengthening visual capabilities beyond speech synthesis technology. A representative example is the recruitment of Chief Research Officer (CRO) Lee Je-hee in April last year. CRO Lee has been a professor in the Department of Computer Science and Engineering at Seoul National University since 2003, and is a scholar in the field of computer graphics and animation. In July, NCsoft recruited Daniel Toshiu as the global art leader and Jeong Byung-gun as the digital actor head. Director Jeong is a digital human expert who worked as a virtual influencer at Sidus Studio X.
https://biz.chosun.com/it-science/ict/2022/09/30/ENZBJ4OHO5GMZLAX66VTL5NJBA/