Generalized End-to-End Loss for Speaker Verification

Abstract

In this paper, we propose a new loss function called generalized end-to-end (GE2E) loss, which makes the training of speaker verification models more efficient than our previous tuple-based end-to-end (TE2E) loss function. Unlike TE2E, the GE2E loss function updates the network in a way that emphasizes examples that are difficult to verify at each step of the training process. Additionally, the GE2E loss does not require an initial stage of example selection. With these properties, our model with the new loss function decreases speaker verification EER by more than 10%, while reducing the training time by 60% at the same time. We also introduce the MultiReader technique, which allows us to do domain adaptation - training a more accurate model that supports multiple keywords (i.e. "OK Google" and "Hey Google") as well as multiple dialects.

arXiv commentPublished at ICASSP 2018

References in corpus (1)

• Deep Speaker: an End-to-End Neural Speaker Embedding System

Cited by in corpus (7)

• Attention-Based Models for Text-Dependent Speaker Verification
• Expressive Neural Voice Cloning
• Links: A High-Dimensional Online Clustering Method
• Replay attack detection with complementary high-resolution information using end-to-end DNN for the ASVspoof 2019 Challenge
• Speaker Embedding Extraction with Phonetic Information
• A study on the role of subsidiary information in replay attack spoofing detection
• Text-Independent Speaker Verification Based on Deep Neural Networks and Segmental Dynamic Time Warping