RQNet: Residual quaternion CNN for performance enhancement in low complexity and device robust acoustic scene classification

A Madhu, K Suresh - IEEE Transactions on Multimedia, 2023 - ieeexplore.ieee.org
Acoustic Scene Classification aims to recognize the unique acoustic characteristics of an
environment. Recently, Convolutional Neural Networks (CNNs) have boosted the accuracy …

Dummy prototypical networks for few-shot open-set keyword spotting

B Kim, S Yang, I Chung, S Chang - arXiv preprint arXiv:2206.13691, 2022 - arxiv.org
Keyword spotting is the task of detecting a keyword in streaming audio. Conventional
keyword spotting targets predefined keywords classification, but there is growing attention in …

[PDF][PDF] Multi-Scale Architecture and Device-Aware Data-Random-Drop Based Fine-Tuning Method for Acoustic Scene Classification.

JH Lee, JH Choi, PM Byun, JH Chang - DCASE, 2022 - dcase.community
We propose a low-complexity acoustic scene classification (ASC) model structure suitable
for short-segmented audio and fine-tuning methods for generalization to multiple recording …

Randmasking augment: A simple and randomized data augmentation for acoustic scene classification

J Han, M Matuszewski, O Sikorski… - ICASSP 2023-2023 …, 2023 - ieeexplore.ieee.org
In this work, we describe RandMasking Augment as an effective data augmentation method
for acoustic scene classification research. We concentrate on both time and frequency …

Efficient Lightweight Speaker Verification With Broadcasting CNN-Transformer and Knowledge Distillation Training of Self-Attention Maps

JH Choi, JY Yang, JH Chang - IEEE/ACM Transactions on …, 2024 - ieeexplore.ieee.org
Developing a lightweight speaker embedding extractor (SEE) is crucial for the practical
implementation of automatic speaker verification (ASV) systems. To this end, we recently …

Domain agnostic few-shot learning for speaker verification

S Yang, D Das, J Cho, H Park, S Yun - arXiv preprint arXiv:2206.13700, 2022 - arxiv.org
Deep learning models for verification systems often fail to generalize to new users and new
environments, even though they learn highly discriminative features. To address this …

Synthetic data generation techniques for training deep acoustic siren identification networks

S Damiano, B Cramer, A Guntoro… - Frontiers in Signal …, 2024 - frontiersin.org
Acoustic sensing has been widely exploited for the early detection of harmful situations in
urban environments: in particular, several siren identification algorithms based on deep …

[PDF][PDF] HYU Submission for The Dcase 2022: Fine-tuning method using device-aware data-random-drop for device-imbalanced acoustic scene classification

JH Lee, JH Choi, PM Byun… - Detection Classif. Acoust …, 2022 - dcase.community
This paper address the Hanyang University team submission for the DCASE 2022
Challenge Low-Complexity Acoustic Scene Classification task. The task aims to design a …

Instance-level loss based multiple-instance learning framework for acoustic scene classification

WG Choi, JH Chang, JM Yang, HG Moon - Applied Acoustics, 2024 - Elsevier
An acoustic scene is inferred by detecting properties combining diverse sounds and
acoustic environments. This study is intended to discover these properties effectively using …

Towards domain generalisation in asr with elitist sampling and ensemble knowledge distillation

R Ahmad, MA Jalal, MU Farooq… - ICASSP 2023-2023 …, 2023 - ieeexplore.ieee.org
Knowledge distillation (KD) has widely been used for model compression and domain
adaptation for speech applications. In the presence of multiple teachers, knowledge can …