Utterance-level Aggregation For Speaker Recognition In The Wild

Weidi Xie, Arsha Nagrani, Joon Son Chung, Andrew Zisserman
2/26/2019
eess.AScs.LGcs.MM

Official code: WeidiXie/VGG-Speaker-Recognition (372 stars, last commit 2023-03-24).

Abstract

The objective of this paper is speaker recognition "in the wild"-where utterances may be of variable length and also contain irrelevant signals. Crucial elements in the design of deep networks for this task are the type of trunk (frame level) network, and the method of temporal aggregation. We propose a powerful speaker recognition deep network, using a "thin-ResNet" trunk architecture, and a dictionary-based NetVLAD or GhostVLAD layer to aggregate features across time, that can be trained end-to-end. We show that our network achieves state of the art performance by a significant margin on the VoxCeleb1 test set for speaker recognition, whilst requiring fewer parameters than previous methods. We also investigate the effect of utterance length on performance, and conclude that for "in the wild" data, a longer length is beneficial.

View on arXivView PDF

Code Implementations(1)

WeidiXie/VGG-Speaker-RecognitionOfficial100%

Utterance-level Aggregation For Speaker Recognition In The Wild

37294PythonFeb 9, 20193 years ago

Similar Papers

Speaker Recognition in Realistic Scenario Using Multimodal Data

84% match2023

Deep Neural Networks for Automatic Speaker Recognition Do Not Learn Supra-Segmental Temporal Features

83% match2023

Combining Residual Networks with LSTMs for Lipreading

83% match1 repo992017

Probing the Information Encoded in Neural-based Acoustic Models of Automatic Speech Recognition Systems

83% match2024

DS-TDNN: Dual-stream Time-delay Neural Network with Global-aware Filter for Speaker Verification

83% match1 repo412023

Enhancing Neural Spoken Language Recognition: An Exploration with Multilingual Datasets

82% match2025

Cite this paper

@article{xie2019utterancelevel,
  title  = {Utterance-level Aggregation For Speaker Recognition In The Wild},
  author = {Weidi Xie and Arsha Nagrani and Joon Son Chung and Andrew Zisserman},
  year   = {2019},
  eprint = {1902.10107},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1902.10107}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1902.10107.svg)](https://codeofpaper.com/paper/1902.10107)

Discussion