Self-supervised learning (SSL) has pushed speaker verification accuracy close to state-of-the-art levels, but the Transformer backbones used in most SSL encoders hinder on-device and real-time deployment. Prior compression work trims layer depth or width yet still inherits the quadratic cost of self-attention. We propose SV-Mixer, the first fully MLPbased student encoder for SSL distillation. SV-Mixer replaces Transformer with three lightweight modules: Multi-Scale Mixing for multi-resolution temporal features, Local-Global Mixing for frame-to-utterance context, and Group Channel Mixing for spectral subspaces. Distilled from WavLM, SV-Mixer outperforms a Transformer student by $14.6 \%$ while cutting parameters and GMACs by over half, and at $\mathbf{7 5 \%}$ compression, it closely matches the teacher’s performance. Our results show that attention-free SSL students can deliver teacher-level accuracy with hardwarefriendly footprints, opening the door to robust on-device speaker verification.

Background noise considerably reduces the accuracy and reliability of speaker verification (SV) systems. These challenges can be addressed using a speech enhancement system as a front-end module. Recently, diffusion probabilistic models (DPMs) have exhibited remarkable noise-compensation capabilities in the speech enhancement domain. Building on this success, we propose Diff-SV, a noise-robust SV framework that leverages DPM. Diff-SV unifies a DPM-based speech enhancement system with a speaker embedding extractor, and yields a discriminative and noise-tolerable speaker representation through a hierarchical structure. The proposed model was evaluated under both in-domain and out-of-domain noisy conditions using the VoxCeleb1 test set, an external noise source, and the VOiCES corpus. The obtained experimental results demonstrate that Diff-SV achieves state-of-the-art performance, outperforming recently proposed noise-robust SV systems.

Audio deepfake detection (ADD) is the task of detecting spoofing attacks generated by text-to-speech or voice conversion systems. Spoofing evidence, which helps to distinguish between spoofed and bona-fide utterances, might exist either locally or globally in the input features. To capture these, the Conformer, which consists of Transformers and CNN, possesses a suitable structure. However, since the Conformer was designed for sequence-to-sequence tasks, its direct application to ADD tasks may be sub-optimal. To tackle this limitation, we propose HM-Conformer by adopting two components: (1) Hierarchical pooling method progressively reducing the sequence length to eliminate duplicated information (2) Multi-level classification token aggregation method utilizing classification tokens to gather information from different blocks. Owing to these components, HM-Conformer can efficiently detect spoofing evidence by processing various sequence lengths and aggregating them. In experimental results on the ASVspoof 2021 Deepfake dataset, HM-Conformer achieved a 15.71% EER, showing competitive performance compared to recent systems.

Speaker verification (SV) technology has the potential to enhance personalization and security in various applications, such as voice assistants, forensics, and access control. However, several challenges hinder the practical application of SV systems, including limitations and distortions in speaker information due to short utterances and noisy environments. Furthermore, these two factors often coexist in real-world situations, resulting in a significant performance degradation of SV systems. Despite the significance of these obstacles, each factor is independently studied, and the co-occurrence of both factors is rarely investigated. Here, we propose a novel SV framework, feature aggregated extended U-Net (FA-ExU-Net), which simultaneously addresses both the challenges by building on the success of prior research on each factor. The FA-ExU-Net incorporates an iterative and hierarchical feature aggregation scheme, a target task-specific feature enhancement module, and a multi-scale feature aggregator for extracting information-rich embeddings. Our proposed system outperforms the recent baseline models based on four evaluation criteria: generalizability, short utterance performance, capacity to handle noisy environments, and robustness to short utterances in noisy environments. We demonstrate the effectiveness of the proposed model through comparison and ablation experiments and intuitive visualizations. The proposed novel approach is expected to contribute to the development of more robust and accurate SV models for practical applications. Our training codes are available athttps://github.com/wngh1187/FA-ExU-Net.

Despite achieving satisfactory performance in speaker verification using deep neural networks, variable-duration utterances remain a challenge that threatens the robustness of systems. To deal with this issue, we propose a speaker verification system called RawNeXt that can handle input raw waveforms of arbitrary length by employing the following two components: (1) A deep layer aggregation strategy enhances speaker information by iteratively and hierarchically aggregating features of various time scales and spectral channels output from blocks. (2) An extended dynamic scaling policy flexibly processes features according to the length of the utterance by selectively merging the activations of different resolution branches in each block. Owing to these two components, our proposed model can extract speaker embeddings rich in time-spectral information and operate dynamically on length variations. Experimental results on the VoxCeleb1 test set consisting of various duration utterances demonstrate that RawNeXt achieves state-of-the-art performance compared to the recently proposed systems. Our code and trained model weights are available at https://github.com/wngh1187/RawNeXt.

The objective of this paper is to combine multiple frame-level features into a single utterance-level representation considering pair-wise relationships. For this purpose, we propose a novel graph attentive feature aggregation module by interpreting each frame-level feature as a node of a graph. The inter-relationship between all possible pairs of features, typically exploited indirectly, can be directly modeled using a graph. The module comprises a graph attention layer and a graph pooling layer followed by a readout operation. The graph attention layer first models the non-Euclidean data manifold between different nodes. Then, the graph pooling layer discards less informative nodes considering the significance of the nodes. Finally, the readout operation combines the remaining nodes into a single representation. We employ two recent systems, SE-ResNet and RawNet2, with different input features and architectures and demonstrate that the proposed feature aggregation module consistently shows a relative improvement over 10%, compared to the baseline.

The use of deep neural networks (DNN) has dramatically elevated the performance of automatic speaker verification (ASV) over the last decade. However, ASV systems can be easily neutralized by spoofing attacks. Therefore, the Spoofing-Aware Speaker Verification (SASV) challenge is designed and held to promote development of systems that can perform ASV considering spoofing attacks by integrating ASV and spoofing countermeasure (CM) systems. In this paper, we propose two back-end systems: multi-layer perceptron score fusion model (MSFM) and integrated embedding projector (IEP). The MSFM, score fusion back-end system, derived SASV score utilizing ASV and CM scores and embeddings. On the other hand,IEP combines ASV and CM embeddings into SASV embedding and calculates final SASV score based on the cosine similarity. We effectively integrated ASV and CM systems through proposed MSFM and IEP and achieved the SASV equal error rates 0.56%, 1.32% on the official evaluation trials of the SASV 2022 challenge.

Background noise is a well-known factor that deteriorates the accuracy and reliability of speaker verification (SV) systems by blurring speech intelligibility.Various studies have used separate pretrained enhancement models as the front-end module of the SV system in noisy environments, and these methods effectively remove noises.However, the denoising process of independent enhancement models not tailored to the SV task can also distort the speaker information included in utterances.We argue that the enhancement network and speaker embedding extractor should be fully jointly trained for SV tasks under noisy conditions to alleviate this issue.Therefore, we proposed a U-Net-based integrated framework that simultaneously optimizes speaker identification and feature enhancement losses.Moreover, we analyzed the structural limitations of using U-Net directly for noise SV tasks and further proposed Extended U-Net to reduce these drawbacks.We evaluated the models on the noise-synthesized VoxCeleb1 test set and VOiCES development set recorded in various noisy scenarios.The experimental results demonstrate that the U-Net-based fully joint training framework is more effective than the baseline, and the extended U-Net exhibited state-of-the-art performance versus the recently proposed compensation systems.