arXiv subjects

Minqiang Xu

Publications and source records attributed to Minqiang Xu.

24 records · Page 2Linked to original sources

arXiv2021

Rep Works in Speaker Verification

Multi-branch convolutional neural network architecture has raised lots of attention in speaker verification since the aggregation of multiple parallel branches can significantly improve performance. However, this design is not efficient enough during the inference time due to the increase of model parameters and extra operations. In this paper, we present a new multi-branch network architecture RepSPKNet that uses a re-parameterization technique. With this technique, our backbone model contains an efficient VGG-like inference state while its training state is a complicated multi-branch structure. We first introduce the specific structure of RepVGG into speaker verification and propose several variants of this structure. The performance is evaluated on VoxCeleb-based test sets. We demonstrate that both the branch diversity and the branch capacity play important roles in RepSPKNet designing. Our RepSPKNet achieves state-of-the-art performance with a 1.5982% EER and a 0.1374 minDCF on VoxCeleb1-H.

cs.SD↗

arXiv2021

Multi-query multi-head attention pooling and Inter-topK penalty for speaker verification

This paper describes the multi-query multi-head attention (MQMHA) pooling and inter-topK penalty methods which were first proposed in our submitted system description for VoxCeleb speaker recognition challenge (VoxSRC) 2021. Most multi-head attention pooling mechanisms either attend to the whole feature through multiple heads or attend to several split parts of the whole feature. Our proposed MQMHA combines both these two mechanisms and gain more diversified information. The margin-based softmax loss functions are commonly adopted to obtain discriminative speaker representations. To further enhance the inter-class discriminability, we propose a method that adds an extra inter-topK penalty on some confused speakers. By adopting both the MQMHA and inter-topK penalty, we achieved state-of-the-art performance in all of the public VoxCeleb test sets.

cs.SD↗

arXiv2021

Poformer: A simple pooling transformer for speaker verification

Most recent speaker verification systems are based on extracting speaker embeddings using a deep neural network. The pooling layer in the network aims to aggregate frame-level features extracted by the backbone. In this paper, we propose a new transformer based pooling structure called PoFormer to enhance the ability of the pooling layer to capture information along the whole time axis. Different from previous works that apply attention mechanism in a simple way or implement the multi-head mechanism in serial instead of in parallel, PoFormer follows the initial transformer structure with some minor modifications like a positional encoding generator, drop path and LayerScale to make the training procedure more stable and to prevent overfitting. Evaluated on various datasets, PoFormer outperforms the existing pooling system with at least a 13.00% improvement in EER and a 9.12% improvement in minDCF.

eess.AS↗

arXiv2021

Fast query-by-example speech search using separable model

Traditional Query-by-Example (QbE) speech search approaches usually use methods based on frame-level features, while state-of-the-art approaches tend to use models based on acoustic word embeddings (AWEs) to transform variable length audio signals into fixed length feature vector representations. However, these approaches cannot meet the requirements of the search quality as well as speed at the same time. In this paper, we propose a novel fast QbE speech search method based on separable models to fix this problem. First, a QbE speech search training framework is introduced. Second, we design a novel model inference scheme based on RepVGG which can efficiently improve the QbE search quality. Third, we modify and improve our QbE speech search model according to the proposed model inference scheme. Experiments on keywords dataset shows that our proposed method can improve the GPU Real-time Factor (RTF) from 1/150 to 1/2300 by just applying separable model scheme and outperforms other state-of-the-art methods.

eess.AS↗

arXiv2021

The SpeakIn System for VoxCeleb Speaker Recognition Challange 2021

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%.

cs.SD↗

arXiv2018

Hessian recovery based finite element methods for the Cahn-Hilliard Equation

In this paper, we propose a novel recovery based finite element method for the Cahn-Hilliard equation. One distinguishing feature of the method is that we discretize the fourth-order differential operator in a standard $C^0$ linear finite elements space. Precisely, we first transform the fourth-order Cahn-Hilliard equation to its variational formulation in which only first-order and second-order derivatives are involved and then we compute the and second-order derivatives of a linear finite element function by a least-squares fitting recovery procedure. The intrinsic link between the second-order derivatives (Hessian matrix) recovery scheme and the finite difference method is studied in the paper. In particular, for the first time, we discover that the Laplace recovery scheme is exactly the well-known five-point stencil over uniform meshes. The proposed discretization for the Cahn-Hilliard equation can be regarded as a combination of the finite difference scheme and the finite element scheme. In addition, special considerations are put on different methods for imposing Neumann type boundary conditions. The optimal-order convergence and energy stability are numerically proved through a series of benchmark tests.

math.NA↗