16 citations · 102 across the 27 of their papers we have counts for
28 papers
PAS: Partial Additive Speech Data Augmentation Method for Noise Robust Speaker Verification
Wonbin Kim, Hyun-seo Shin, Ju-ho Kim +3
Background noise reduces speech intelligibility and quality, making speaker verification (SV) in noisy environments a challenging task. To improve the noise robustness of SV system…
Convolution channel separation and frequency sub-bands aggregation for music genre classification
Jungwoo Heo, Hyun-seo Shin, Ju-ho Kim +2
In music, short-term features such as pitch and tempo constitute long-term semantic features such as melody and narrative. A music genre classification (MGC) system should be able…
Integrated Parameter-Efficient Tuning for General-Purpose Audio Models
Ju-ho Kim, Jungwoo Heo, Hyun-seo Shin +2
The advent of hyper-scale and general-purpose pre-trained models is shifting the paradigm of building task-specific models for target tasks. In the field of audio research, task-ag…
Extended U-Net for Speaker Verification in Noisy Environments
Ju-ho Kim, Jungwoo Heo, Hye-jin Shim +1
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…
Baseline Systems for the First Spoofing-Aware Speaker Verification Challenge: Score and Embedding Fusion
Hye-jin Shim, Hemlata Tak, Xuechen Liu +12
Deep learning has brought impressive progress in the study of both automatic speaker verification (ASV) and spoofing countermeasures (CM). Although solutions are mutually dependent…
SASV 2022: The First Spoofing-Aware Speaker Verification Challenge
Jee-weon Jung, Hemlata Tak, Hye-jin Shim +6
The first spoofing-aware speaker verification (SASV) challenge aims to integrate research efforts in speaker verification and anti-spoofing. We extend the speaker verification scen…
SASV Challenge 2022: A Spoofing Aware Speaker Verification Challenge Evaluation Plan
Jee-weon Jung, Hemlata Tak, Hye-jin Shim +7
ASV (automatic speaker verification) systems are intrinsically required to reject both non-target (e.g., voice uttered by different speaker) and spoofed (e.g., synthesised or conve…
Graph attentive feature aggregation for text-independent speaker verification
Hye-jin Shim, Jungwoo Heo, Jae-han Park +2
The objective of this paper is to combine multiple frame-level features into a single utterance-level representation considering pairwise relationship. For this purpose, we propose…
RawNeXt: Speaker verification system for variable-duration utterances with deep layer aggregation and extended dynamic scaling policies
Ju-ho Kim, Hye-jin Shim, Jungwoo Heo +1
Despite achieving satisfactory performance in speaker verification using deep neural networks, variable-duration utterances remain a challenge that threatens the robustness of syst…
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
Jee-weon Jung, Hee-Soo Heo, Hemlata Tak +5
Artefacts that differentiate spoofed from bona-fide utterances can reside in spectral or temporal domains. Their reliable detection usually depends upon computationally demanding e…
Learning Metrics from Mean Teacher: A Supervised Learning Method for Improving the Generalization of Speaker Verification System
Ju-ho Kim, Hye-jin Shim, Jee-weon Jung +1
Most speaker verification tasks are studied as an open-set evaluation scenario considering the real-world condition. Thus, the generalization power to unseen speakers is of paramou…
Attentive max feature map and joint training for acoustic scene classification
Hye-jin Shim, Jee-weon Jung, Ju-ho Kim +1
Various attention mechanisms are being widely applied to acoustic scene classification. However, we empirically found that the attention mechanism can excessively discard potential…
Graph Attention Networks for Speaker Verification
Jee-weon Jung, Hee-Soo Heo, Ha-Jin Yu +1
This work presents a novel back-end framework for speaker verification using graph attention networks. Segment-wise speaker embeddings extracted from multiple crops within an utter…
DcaseNet: An integrated pretrained deep neural network for detecting and classifying acoustic scenes and events
Jee-weon Jung, Hye-jin Shim, Ju-ho Kim +1
Although acoustic scenes and events include many related tasks, their combined detection and classification have been scarcely investigated. We propose three architectures of deep…
Capturing scattered discriminative information using a deep architecture in acoustic scene classification
Hye-jin Shim, Jee-weon Jung, Ju-ho Kim +1
Frequently misclassified pairs of classes that share many common acoustic properties exist in acoustic scene classification (ASC). To distinguish such pairs of classes, trivial det…
Integrated Replay Spoofing-aware Text-independent Speaker Verification
Hye-jin Shim, Jee-weon Jung, Ju-ho Kim +2
A number of studies have successfully developed speaker verification or presentation attack detection systems. However, studies integrating the two tasks remain in the preliminary…
Improved RawNet with Feature Map Scaling for Text-independent Speaker Verification using Raw Waveforms
Jee-weon Jung, Seung-bin Kim, Hye-jin Shim +2
Recent advances in deep learning have facilitated the design of speaker verification systems that directly input raw waveforms. For example, RawNet extracts speaker embeddings from…
Segment Aggregation for short utterances speaker verification using raw waveforms
Seung-bin Kim, Jee-weon Jung, Hye-jin Shim +2
Most studies on speaker verification systems focus on long-duration utterances, which are composed of sufficient phonetic information. However, the performances of these systems ar…
Acoustic Scene Classification using Audio Tagging
Jee-weon Jung, Hye-jin Shim, Ju-ho Kim +2
Acoustic scene classification systems using deep neural networks classify given recordings into pre-defined classes. In this study, we propose a novel scheme for acoustic scene cla…
A study on the role of subsidiary information in replay attack spoofing detection
Jee-weon Jung, Hye-jin Shim, Hee-Soo Heo +1
In this study, we analyze the role of various categories of subsidiary information in conducting replay attack spoofing detection: `Room Size', `Reverberation', `Speaker-to-ASV dis…
Self-supervised pre-training with acoustic configurations for replay spoofing detection
Hye-jin Shim, Hee-Soo Heo, Jee-weon Jung +1
Constructing a dataset for replay spoofing detection requires a physical process of playing an utterance and re-recording it, presenting a challenge to the collection of large-scal…
Cosine similarity-based adversarial process
Hee-Soo Heo, Jee-weon Jung, Hye-jin Shim +2
An adversarial process between two deep neural networks is a promising approach to train a robust model. In this paper, we propose an adversarial process using cosine similarity, w…
Acoustic scene classification using teacher-student learning with soft-labels
Hee-Soo Heo, Jee-weon Jung, Hye-jin Shim +1
Acoustic scene classification identifies an input segment into one of the pre-defined classes using spectral information. The spectral information of acoustic scenes may not be mut…
Replay attack detection with complementary high-resolution information using end-to-end DNN for the ASVspoof 2019 Challenge
Jee-weon Jung, Hye-jin Shim, Hee-Soo Heo +1
In this study, we concentrate on replacing the process of extracting hand-crafted acoustic feature with end-to-end DNN using complementary high-resolution spectrograms. As a result…