NAVER Presents 8 Research Papers at INTERSPEECH 2022…Showcasing World-Class Voice AI Technologies
NAVER Presents 8 Research Papers at INTERSPEECH 2022…Showcasing World-Class Voice AI Technologies
NAVER Presents 8 Research Papers at INTERSPEECH 2022…Showcasing World-Class Voice AI Technologies
- 12 papers published in total when taking LINE into account…recognition of the company’s technological capabilities by presenting papers at INTERSPEECH for 5 consecutive years
- Introduced multiple voice AI technologies that could be applied to NAVER’s services including CLOVA Dubbing and CLOVA Note
- Hosted a booth dedicated to NAVER and “NAVER Night” to attract global talents
26th Sep. 2022
NAVER Corp. (KRX: 035420) has presented 8 research papers at “INTERSPEECH 2022,” the world’s most renowned academic conference for speech signal processing, being recognized for its world-class voice AI technologies.
INTERSPEECH, which has marked its 23rd anniversary this year, is one of the most prestigious academic conferences in the field of voice AI technology along with ICASSP. INTERSPEECH is where thousands of experts in voice AI technology share the results of cutting-edge research. This year’s event was held from September 18 to 22 at Songdo ConvensiA in Incheon, Korea.
NAVER has presented 8 papers in total this year with 5 of them being selected for oral presentation. The total number of presented papers amounts to 12 when also including the papers presented by LINE. Papers on joint research conducted with prestigious institutions both in and out of Korea, including Yonsei University, Carnegie Mellon University, and EURECOM, were also presented.
In particular, not only mid- to long-term advanced R&D, but also R&D on practical technologies that are likely to be applied to NAVER’s services were showcased, including applying emotions to speech synthesis, improving speaker recognition rate, and enhancing memory efficiency when training speech recognition models. These technologies are embedded into CLOVA AI speaker, CLOVA Dubbing, CLOVA Note, and CLOVA CareCall.
In addition to presenting research papers at INTERSPEECH 2022, NAVER proactively participated in many sessions as one of the leading organizations in the field of voice AI. NAVER’s researchers led the SASV (Spoofing-Aware Automatic Speaker Verification) 2022 Challenge, a special session for ASV (automatic speaker verification), while Lee Chan-kyu, the leader for developing speech models at CLOVA, participated as a speaker at Industry Talk and introduced CLOVA’s speech recognition technology and service to world-leading AI researchers.
NAVER, a diamond sponsor of INTERSPEECH 2022, hosted a booth to showcase the latest AI technologies and to recruit global top talents. Meanwhile, about 60 student researchers attended the “NAVER Night” event held on the 19th and spent time sharing ideas and thoughts with NAVER’s researchers.
Jeong Seok-keun, Head of NAVER CLOVA said, “NAVER has been recognized for its outstanding voice AI capabilities at INTERSPEECH for 5 consecutive years. With such capabilities, we were capable of providing innovative AI services to our users. We plan to continue investing in the advancement of voice AI technology and to further solidify AI leadership going forward.”
NAVER has topped last year’s number of 66 papers by already publishing 88 research papers in top-tier AI conferences including INTERSPEECH, ICML, CVPR, ECCV, and NAACL in the course of this year.
<END>