TristouNet: Triplet Loss for Speaker Turn Embedding

ABSTRACT TristouNet is a neural network architecture based on Long Short-Term Memory recurrent networks, meant to project speech sequences into a fixed-dimensional euclidean space. Thanks to the triplet loss paradigm used for training, the resulting sequence embeddings can be compared directly with the euclidean distance, for speaker comparison purposes. Experiments on short (between 500ms and 5s) speech turn comparison and speaker change detection show that TristouNet brings significant improvements over the current state-of-the-art techniques for both tasks. Index Terms-- triplet loss, long short-term memory network, sequence embedding, speaker recognition 1. INTRODUCTION Given a speech sequence x and a claimed identity a, speaker verification aims at accepting or rejecting the identity claim. Speaker identification is the task of determining which speaker (from a predefined set of speakers a S) has uttered the sequence x.