A Novel Speaker Binary Key Derived from Anchor Models

摘要: The approach presented in this paper represents voice recordings by a novel acoustic key composed only of binary values. Except for the process being used to extract such keys, there is no need modeling and processing proposed, as all other elements system are based on vectors. We show that able effectively model speaker’s distinguish it from speakers. Its main properties its small size compared current speaker techniques low computational cost when comparing different speakers limited obtaining similarity metric between two Furthermore, vector extraction does not any threshold offers opportunity set decision steps well defined domain where scores decisions easy interpret implement. Index Terms: key, modeling, biometrics