Abstract: A circuit board according to an embodiment includes an insulating layer; a first circuit pattern disposed on a first surface of the insulating layer; a first solder resist disposed on the first surface of the insulating layer; and a first barrier layer including a first-first portion disposed between the first solder resist and the first circuit pattern, and a first-second portion disposed between the insulating layer and the first circuit pattern; wherein the firs-first portion of the first barrier layer includes: a first-first gold (Au) layer disposed under a lower surface of the first circuit pattern; and a first-first palladium (Pd) layer disposed under a lower surface of the first-first gold (Au) layer; wherein the first-second portion of the first barrier layer includes: a first-second gold (Au) layer disposed to surround a side surface and an upper surface of the first circuit pattern; and a first-second palladium (Pd) layer disposed to surround the first-second gold (Au) layer; and wherein the first c

Abstract: The present invention relates to a speaker diarization technology, and more specifically to, end-to-end speaker diarization system and method through transformer learning having an auxiliary loss-based residual connection to separate speakers by dividing the speakers for time interval, wherein the end-to-end speaker diarization system and method using an auxiliary loss can differentiate and separate speakers through speaker labeling based on the transformer learning using an auxiliary loss even if speaker speeches overlap in a multi-speaker environment.