I4U Submission to NIST SRE 2018: Leveraging from a Decade of Shared Experiences

• Main Authors: Lee, Kong Aik, Hautamaki, Ville, Kinnunen, Tomi, Yamamoto, Hitoshi, Okabe, Koji, Vestman, Ville, Huang, Jing, Ding, Guohong, Sun, Hanwu, Larcher, Anthony, Das, Rohan Kumar, Li, Haizhou, Rouvier, Mickael, Bousquet, Pierre-Michel, Rao, Wei, Wang, Qing, Zhang, Chunlei, Bahmaninezhad, Fahimeh, Delgado, Hector, Patino, Jose, Wang, Qiongqiong, Guo, Ling, Koshinaka, Takafumi, Zhang, Jiacen, Shinoda, Koichi, Trong, Trung Ngo, Sahidullah, Md, Lu, Fan, Tang, Yun, Tu, Ming, Teh, Kah Kuan, Tran, Huy Dat, George, Kuruvachan K, Kukanov, Ivan, Desnous, Florent, Yang, Jichen, Yilmaz, Emre, Xu, Longting, Bonastre, Jean-Francois, Xu, Chenglin, Lim, Zhi Hao, Chng, Eng Siong, Ranjan, Shivesh, Hansen, John H. L, Todisco, Massimiliano, Evans, Nicholas
• Format: Journal Article
• Language: English
• Published: 15.04.2019
• Subjects: Computer Science - Computation and Language; Computer Science - Sound
• DOI: 10.48550/arxiv.1904.07386

The I4U consortium was established to facilitate a joint entry to NIST speaker recognition evaluations (SRE). The latest edition of such joint submission was in SRE 2018, in which the I4U submission was among the best-performing systems. SRE'18 also marks the 10-year anniversary of I4U consortium into NIST SRE series of evaluation. The primary objective of the current paper is to summarize the results and lessons learned based on the twelve sub-systems and their fusion submitted to SRE'18. It is also our intention to present a shared view on the advancements, progresses, and major paradigm shifts that we have witnessed as an SRE participant in the past decade from SRE'08 to SRE'18. In this regard, we have seen, among others, a paradigm shift from supervector representation to deep speaker embedding, and a switch of research challenge from channel compensation to domain adaptation.