This technical report describes the SJTU X-LANCE Lab system for the three tracks in CNSRC 2022. In this challenge, we explored the speaker embedding modeling ability of deep ResNet (Deeper r-vector). All the systems are only trained on the Cnceleb training set and we use the same systems for the three tracks in CNSRC 2022. In this challenge, our system ranks the first place in the fixed track of speaker verification task. Our best single system and fusion system achieve 0.3164 and 0.2975 minDCF respectively. Besides, we submit the result of ResNet221 to the speaker retrieval track and achieve 0.4626 mAP. More importantly, we have helped the wespeaker [1] toolkit reproduce our result: https://github.com/wenet-e2e/wespeaker.

The SJTU X-LANCE Lab System for CNSRC 2022

Embeddings extracted by deep neural networks have become the state-of-the-art utterance representation in speaker verification (SV). Despite the various network architectures that have been investigated in previous works, how to design and scale up networks to achieve a better trade-off on performance and complexity in a principled manner has been rarely discussed in the SV field. In this paper, we first systematically study model scaling from the perspective of the depth and width of networks and empirically discover that depth is more important than the width of networks for speaker verification task . Based on this observation, we design a new backbone constructed en-tirely from standard convolutional network modules by significantly increasing the number of layers while maintaining the network complexity following the depth-first rule and scale it up to obtain a family of much deeper models dubbed DF-ResNets. Comprehensive comparisons with other state-of-the-art systems on the Voxceleb dataset demonstrate that DF-ResNets achieve a much better trade-off than previous SV systems in terms of performance and complexity.

DF-ResNet: Boosting Speaker Verification Performance with Depth-First Design

Speaker modeling is essential for many related tasks, such as speaker recognition and speaker diarization. The dominant modeling approach is fixed-dimensional vector representation, i.e., speaker embedding. This paper introduces a research and production oriented speaker embedding learning toolkit, Wespeaker. Wespeaker contains the implementation of scalable data management, state-of-the-art speaker embedding models, loss functions, and scoring back-ends, with highly competitive results achieved by structured recipes which were adopted in the winning systems in several speaker verification challenges. The application to other downstream tasks such as speaker diarization is also exhibited in the related recipe. Moreover, CPU- and GPU-compatible deployment codes are integrated for production-oriented development. The toolkit is publicly available at https://github.com/wenet-e2e/wespeaker.

Wespeaker: A Research and Production oriented Speaker Embedding Learning Toolkit

Many speaker recognition challenges have been held to assess the speaker verification system in the wild and probe the performance limit. Voxceleb Speaker Recognition Challenge (VoxSRC), based on the voxceleb, is the most popular. Besides, another challenge called CN-Celeb Speaker Recognition Challenge (CNSRC) is also held this year, which is based on the Chinese celebrity multi-genre dataset CN-Celeb. This year, our team participated in both speaker verification closed tracks in CNSRC 2022 and VoxSRC 2022, and achieved the 1st place and 3rd place respectively. In most system reports, the authors usually only provide a description of their systems but lack an effective analysis of their methods. In this paper, we will outline how to build a strong speaker verification challenge system and give a detailed analysis of each method compared with some other popular technical means.

Build a SRE Challenge System: Lessons from VoxSRC 2022 and CNSRC 2022

In this paper, we propose a novel self-distillation method for fake speech detection (FSD), which can significantly improve the performance of FSD without increasing the model complexity. For FSD, some fine-grained information is very important, such as spectrogram defects, mute segments, and so on, which are often perceived by shallow networks. However, shallow networks have much noise, which can not capture this very well. To address this problem, we propose using the deepest network instruct shallow network for enhancing shallow networks. Specifically, the networks of FSD are divided into several segments, the deepest network being used as the teacher model, and all shallow networks become multiple student models by adding classifiers. Meanwhile, the distillation path between the deepest network feature and shallow network features is used to reduce the feature difference. A series of experimental results on the ASVspoof 2019 LA and PA datasets show the effectiveness of the proposed method, with significant improvements compared to the baseline.

Learning From Yourself: A Self-Distillation Method for Fake Speech Detection

As the most widely used technique, deep speaker embedding learning has become predominant in speaker verification task recently. Very large neural networks such as ECAPA-TDNN and ResNet can achieve the state-of-the-art performance. However, large models are computationally unfriendly in general, which require massive storage and computation resources. Model compression has been a hot research topic. Parameter quantization usually results in significant performance degradation. Knowledge distillation demands a pretrained complex teacher model. In this paper, we introduce a novel self-knowledge distillation method, namely Self-Knowledge Distillation via Feature Enhancement (SKDFE). It utilizes an auxiliary self-teacher network to distill its own refined knowledge without the need of a pretrained teacher network. Additionally, we apply the self-knowledge distillation at two different levels: label level and feature level. Experiments on Voxceleb dataset show that our proposed self-knowledge distillation method can make small models have comparable or even better performance than large ones. Large models can also be further improved when applying our method.

Self-Knowledge Distillation via Feature Enhancement for Speaker Verification

Convolution and self-attention based neural networks have both obtained excellent performance in automatic speaker verification. However, the convolution model often lacks the ability of long-term dependency modeling due to the limitation of receptive field, while the self-attention model is insufficient to model local information. To tackle this limitation, we propose a new multi-layer perceptrons based speaker verification network (MLP-SVNet) which can apply MLPs across temporal and frequency dimensions to capture the local and global information at the same time. The experimental results conducted on Voxceleb show that the proposed model is very competitive when compared to other systems based on convolution or self-attention. In addition, we demonstrate that MLP-SVNet based on multi-layer per-ceptrons can produce complementary embeddings, which can be fused with the state-of-the-art system to further improve the performance.

MLP-SVNET: A Multi-Layer Perceptrons Based Network for Speaker Verification

This report describes the SJTU-AISPEECH system for the Voxceleb Speaker Recognition Challenge 2022. For track1, we implemented two kinds of systems, the online system and the ofﬂine system. Different ResNet-based backbones and loss functions are explored. Our ﬁnal fusion system achieved 3rd place in track1. For track3, we implemented statistic adaptation and jointly training based domain adaptation. In the jointly training based domain adaptation, we jointly trained the source and target domain dataset with different training objectives to do the domain adaptation. We explored two different training objectives for target domain data, self-supervised learning based angular proto-typical loss and semi-supervised learning based classiﬁcation loss with estimated pseudo labels. Besides, we used the dynamic loss-gate and label correction (DLG-LC) strategy to improve the quality of pseudo labels when the target domain objective is a classiﬁcation loss. Our ﬁnal fusion system achieved 4th place (very close to 3rd place, relatively less than 1%) in track3.

Bei Liu

Papers

Self-Knowledge Distillation via Feature Enhancement for Speaker Verification

MLP-SVNET: A Multi-Layer Perceptrons Based Network for Speaker Verification

The SJTU X-LANCE Lab System for CNSRC 2022

DF-ResNet: Boosting Speaker Verification Performance with Depth-First Design

SJTU-AISPEECH System for VoxCeleb Speaker Recognition Challenge 2022