303 citations · 697 across the 37 of their papers we have counts for
17 papers · 1 filter
The USTC-NERCSLIP Systems for the CHiME-7 DASR Challenge
Ruoyu Wang, Maokui He, Jun Du +16
This technical report details our submission system to the CHiME-7 DASR Challenge, which focuses on speaker diarization and speech recognition under complex multi-speaker scenarios…
Semi-supervised multi-channel speaker diarization with cross-channel attention
Shilong Wu, Jun Du, Maokui He +4
Most neural speaker diarization systems rely on sufficient manual training data labels, which are hard to collect under real-world scenarios. This paper proposes a semi-supervised…
Variance-Preserving-Based Interpolation Diffusion Models for Speech Enhancement
Zilu Guo, Jun Du, Chin-Hui Lee +2
The goal of this study is to implement diffusion models for speech enhancement (SE). The first step is to emphasize the theoretical foundation of variance-preserving (VP)-based int…
A Multi-dimensional Deep Structured State Space Approach to Speech Enhancement Using Small-footprint Models
Pin-Jui Ku, Chao-Han Huck Yang, Sabato Marco Siniscalchi +1
We propose a multi-dimensional structured state space (S4) approach to speech enhancement. To better capture the spectral dependencies across the frequency axis, we focus on modify…
An Ensemble Teacher-Student Learning Approach with Poisson Sub-sampling to Differential Privacy Preserving Speech Recognition
Chao-Han Huck Yang, Jun Qi, Sabato Marco Siniscalchi +1
We propose an ensemble learning framework with Poisson sub-sampling to effectively train a collection of teacher models to issue some differential privacy (DP) guarantee for traini…
The USTC-Ximalaya system for the ICASSP 2022 multi-channel multi-party meeting transcription (M2MeT) challenge
Maokui He, Xiang Lv, Weilin Zhou +8
We propose two improvements to target-speaker voice activity detection (TS-VAD), the core component in our proposed speaker diarization system that was submitted to the 2022 Multi-…
A Variational Bayesian Approach to Learning Latent Variables for Acoustic Knowledge Transfer
Hu Hu, Sabato Marco Siniscalchi, Chao-Han Huck Yang +1
We propose a variational Bayesian (VB) approach to learning distributions of latent variables in deep neural network (DNN) models for cross-domain knowledge transfer, to address ac…
Separation Guided Speaker Diarization in Realistic Mismatched Conditions
Shu-Tong Niu, Jun Du, Lei Sun +1
We propose a separation guided speaker diarization (SGSD) approach by fully utilizing a complementarity of speech separation and speaker clustering. Since the conventional clusteri…
Device-Robust Acoustic Scene Classification Based on Two-Stage Categorization and Data Augmentation
Hu Hu, Chao-Han Huck Yang, Xianjun Xia +13
In this technical report, we present a joint effort of four groups, namely GT, USTC, Tencent, and UKE, to tackle Task 1 - Acoustic Scene Classification (ASC) in the DCASE 2020 Chal…
On Mean Absolute Error for Deep Neural Network Based Vector-to-Vector Regression
Jun Qi, Jun Du, Sabato Marco Siniscalchi +2
In this paper, we exploit the properties of mean absolute error (MAE) as a loss function for the deep neural network (DNN) based vector-to-vector regression. The goal of this work…
Exploring Deep Hybrid Tensor-to-Vector Network Architectures for Regression Based Speech Enhancement
Jun Qi, Hu Hu, Yannan Wang +3
This paper investigates different trade-offs between the number of model parameters and enhanced speech qualities by employing several deep tensor-to-vector regression models for s…
Relational Teacher Student Learning with Neural Label Embedding for Device Adaptation in Acoustic Scene Classification
Hu Hu, Sabato Marco Siniscalchi, Yannan Wang +1
In this paper, we propose a domain adaptation framework to address the device mismatch issue in acoustic scene classification leveraging upon neural label embedding (NLE) and relat…
An Acoustic Segment Model Based Segment Unit Selection Approach to Acoustic Scene Classification with Partial Utterances
Hu Hu, Sabato Marco Siniscalchi, Yannan Wang +3
In this paper, we propose a sub-utterance unit selection framework to remove acoustic segments in audio recordings that carry little information for acoustic scene classification (…
L-Vector: Neural Label Embedding for Domain Adaptation
Zhong Meng, Hu Hu, Jinyu Li +4
We propose a novel neural label embedding (NLE) scheme for the domain adaptation of a deep neural network (DNN) acoustic model with unpaired data samples from source and target dom…
Characterizing Speech Adversarial Examples Using Self-Attention U-Net Enhancement
Chao-Han Huck Yang, Jun Qi, Pin-Yu Chen +2
Recent studies have highlighted adversarial examples as ubiquitous threats to the deep neural network (DNN) based speech recognition systems. In this work, we present a U-Net based…
Tensor-to-Vector Regression for Multi-channel Speech Enhancement based on Tensor-Train Network
Jun Qi, Hu Hu, Yannan Wang +3
We propose a tensor-to-vector regression approach to multi-channel speech enhancement in order to address the issue of input size explosion and hidden-layer size expansion. The key…
Acoustics-guided evaluation (AGE): a new measure for estimating performance of speech enhancement algorithms for robust ASR
Li Chai, Jun Du, Chin-Hui Lee
One challenging problem of robust automatic speech recognition (ASR) is how to measure the goodness of a speech enhancement algorithm (SEA) without calculating the word error rate…