The SpeakIn System for VoxCeleb Speaker Recognition Challange 2021
arXiv:2109.01989
Abstract
This report describes our submission to the track 1 and track 2 of the VoxCeleb Speaker Recognition Challenge 2021 (VoxSRC 2021). Both track 1 and track 2 share the same speaker verification system, which only uses VoxCeleb2-dev as our training set. This report explores several parts, including data augmentation, network structures, domain-based large margin fine-tuning, and back-end refinement. Our system is a fusion of 9 models and achieves first place in these two tracks of VoxSRC 2021. The minDCF of our submission is 0.1034, and the corresponding EER is 1.8460%.
arXiv commentSubmitted to INTERSPEECH2021 VoxSRC2021 Workshop
References in corpus (10)
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- VoxCeleb2: Deep Speaker Recognition
- VoxCeleb: a large-scale speaker identification dataset
- Attentive Statistics Pooling for Deep Speaker Embedding
- SpeechBrain: A General-Purpose Speech Toolkit
- BUT System Description to VoxCeleb Speaker Recognition Challenge 2019
- The IDLAB VoxSRC-20 Submission: Large Margin Fine-Tuning and Quality-Aware Score Calibration in DNN Based Speaker Verification
- The DKU-DukeECE Systems for VoxCeleb Speaker Recognition Challenge 2020
- Large Margin Softmax Loss for Speaker Verification
- The IDLAB VoxCeleb Speaker Recognition Challenge 2020 System Description
Cited by in corpus (5)
- WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing
- The VoxCeleb Speaker Recognition Challenge: A Retrospective
- Tackling the Score Shift in Cross-Lingual Speaker Verification by Exploiting Language Information
- Rep Works in Speaker Verification
- Poformer: A simple pooling transformer for speaker verification