| Title: |
The I4U Mega Fusion and Collaboration for NIST Speaker Recognition Evaluation 2016 |
| Authors: |
Lee, K; Hautamäki, V; Kinnunen, T; Larcher, A; Zhang, C; Nautsch, A; Stafylakis, T; LIU, G; Rouvier, M; Rao, W; Alegre, F; Ma, J; Mak, M; Sarkar, A; Delgado, H; Saeidi, R; Aronowitz, H; Sizov, A; Sun, H; Nguyen, T; Wang, G; Ma, B; Vestman, V; Sahidullah, M; Halonen, M; Kanervisto, A; Le Lan, G; Bahmaninezhad, F; Isadskiy, S; Rathgeb, C; Busch, C; Tzimiropoulos, G; Qian, Q; Wang, Z; Zhao, Q; Wang, T.; Li, H; Xue, J; Zhu, S; Jin, R; Zhao, T; Bousquet, P.-M; Ajili, M; Kheder, W; Matrouf, D; Lim, Z; Xu, C; Xu, H; Xiao, X; Chng, E; Fauve, B; Sriskandaraja, K; Sethu, V; Lin, W; Thomsen, D; Tan, Z.-H; Todisco, M; Evans, N; Hansen, J; Bonastre, J.-F; Ambikairajah, E |
| Contributors: |
Physikalisches Institut Bonn; Rheinische Friedrich-Wilhelms-Universität Bonn |
| Source: |
Annual Conference of the International Association of Speech Communication (Interspeech) ; https://hal.archives-ouvertes.fr/hal-01927561 ; Annual Conference of the International Association of Speech Communication (Interspeech), Aug 2017, Stockholm, Sweden |
| Publisher Information: |
HAL CCSD |
| Publication Year: |
2017 |
| Collection: |
Archive ouverte HAL (Hyper Article en Ligne, CCSD - Centre pour la Communication Scientifique Directe) |
| Subject Terms: |
bench-mark; Call My Net; Index Terms: speaker recognition evaluation; fusion; [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]; [INFO]Computer Science [cs] |
| Subject Geographic: |
Stockholm; Sweden |
| Time: |
Stockholm, Sweden |
| Description: |
International audience ; The 2016 speaker recognition evaluation (SRE'16) is the latest edition in the series of benchmarking events conducted by the National Institute of Standards and Technology (NIST). I4U is a joint entry to SRE'16 as the result from the collaboration and active exchange of information among researchers from sixteen Institutes and Universities across 4 continents. The joint submission and several of its 32 subsystems were among top-performing systems. A lot of efforts have been devoted to two major challenges, namely, unlabeled training data and dataset shift from Switchboard-Mixer to the new Call My Net dataset. This paper summarizes the lessons learned, presents our shared view from the sixteen research groups on recent advances, major paradigm shift, and common tool chain used in speaker recognition as we have witnessed in SRE'16. More importantly, we look into the intriguing question of fusing a large ensemble of subsystems and the potential benefit of large-scale collaboration . |
| Document Type: |
conference object |
| Language: |
English |
| Relation: |
hal-01927561; https://hal.archives-ouvertes.fr/hal-01927561; https://hal.archives-ouvertes.fr/hal-01927561/document; https://hal.archives-ouvertes.fr/hal-01927561/file/Mega_Fusion_I4U.pdf |
| Availability: |
https://hal.archives-ouvertes.fr/hal-01927561; https://hal.archives-ouvertes.fr/hal-01927561/document; https://hal.archives-ouvertes.fr/hal-01927561/file/Mega_Fusion_I4U.pdf |
| Rights: |
info:eu-repo/semantics/OpenAccess |
| Accession Number: |
edsbas.43CB97B3 |
| Database: |
BASE |