1 Self-Supervised Training of Speaker Encoder with Multi-Modal Diverse Positive Pairs
1Self-SupervisedTrainingofSpeakerEncoderwithMulti-ModalDiversePositivePairsRuijieTao,StudentMember,IEEE,,KongAikLee,SeniorMember,IEEE,RohanKumarDas,SeniorMember,IEEE,VilleHautam¨aki,Member,IEEE,andHaizhouLi,Fellow,IEEEAbstractWestudyanovelneuralarchitectureanditstrainingstrategiesofspeakerencoderfo...
2025-04-30
1.9MB 13 页 0
0
10玖币
渝公网安备50010702506394 | 违法与不良信息举报方式：微信:jiubeiyun2024,QQ:264159069,电话:15523442343,邮箱:jiubeiyun@126.com
