Lightweight Speaker Verification for Online Identification of New Speakers with Short Segments

Verifying if two audio segments belong to the same speaker has been recently put forward as a flexible way to carry out speaker identification, since it does not require to be re-trained when new speakers appear on the auditory scene. However, many of the current techniques employ a considerably high amount of memory, and require a specific minimum audio segment length to obtain good performances. This limits the applicability in areas such as service robots, internet of things and virtual assistants. In this work we propose a BLSTM-based model that reaches a level of performance comparable to the current state of the art when using short input audio segments, while requiring a considerably less amount of memory. Further, as far as we know, a complete speaker identification system has not been reported using this verification paradigm. Thus, we present a complete online speaker identifier, based on a simple voting system that shows that the proposed BLSTM-based model and the current state of the art are similarly accurate at identifying speakers online. \end{abstract}