Search Results for author:
Found 38 papers, 6 papers with code
Eliminating Quantization Errors in Classification-Based Sound Source Localization
To address this, we propose an Unbiased Label Distribution (ULD) to eliminate quantization error in training targets.
Diffusion-Based Adversarial Purification for Speaker Verification
Recently, automatic speaker verification (ASV) based on deep learning is easily contaminated by adversarial attacks, which is a new type of attack that injects imperceptible perturbations to audio signals so as to make ASV produce wrong decisions.
Soft Label Coding for End-to-end Sound Source Localization With Ad-hoc Microphone Arrays
The core idea is to take the geometric connection between the classes into the label coding process. The first one is named static soft label coding (SSLC), which modifies the one-hot codes into soft codes based on the distances between the local areas.
Optimizing Quantum Federated Learning Based on Federated Quantum Natural Gradient Descent
In this work, we propose an efficient optimization algorithm, namely federated quantum natural gradient descent (FQNGD), and further, apply it to a QFL framework that is composed of a variational quantum circuit (VQC)-based quantum neural networks (QNN).
Interpretable Spectrum Transformation Attacks to Speaker Recognition
Unlike existing approaches that operate voices in the time domain, the proposed framework operates voices in the time-frequency domain, which improves the interpretability, transferability, and imperceptibility of the attack.
LMD: A Learnable Mask Network to Detect Adversarial Examples for Speaker Verification
It utilizes score variation as an indicator to detect adversarial examples, where the score variation is the absolute discrepancy between the ASV scores of an original audio recording and its transformed audio synthesized from its masked complex spectrogram.
WeKws: A production first small-footprint end-to-end Keyword Spotting Toolkit
Keyword spotting (KWS) enables speech-based user interaction and gradually becomes an indispensable component of smart devices.
Symmetric Saliency-based Adversarial Attack To Speaker Identification
Adversarial attack approaches to speaker identification either need high computational cost or are not very effective, to our knowledge.
Deep Learning Based Stage-wise Two-dimensional Speaker Localization with Large Ad-hoc Microphone Arrays
To further boost the estimation accuracy, we introduce a node selection algorithm that strategically filters the most reliable nodes.
End-to-end Two-dimensional Sound Source Localization With Ad-hoc Microphone Arrays
Accordingly, the sound source localization problem is formulated as such a classification task of recognizing the one-hot code of the speaker given the one hot codes of the microphone nodes and their speech recordings.
Improving Pseudo Labels With Intra-Class Similarity for Unsupervised Domain Adaptation
In this paper, we propose a novel approach to improve the accuracy of the pseudo labels in the target domain.
Conformer-based End-to-end Speech Recognition With Rotary Position Embedding
To make use of the time order of the input sequence, many works inject some information about the relative or absolute position of the element into the input sequence.
AUC Optimization for Robust Small-footprint Keyword Spotting with Limited Training Data
Deep neural networks provide effective solutions to small-footprint keyword spotting (KWS).
fMBN-E: Efficient Unsupervised Network Structure Ensemble and Selection for Clustering
Empirically, comparing to a number of advanced deep clustering methods and as many as 20 representative unsupervised ensemble learning and selection methods, the proposed methods reach the state-of-the-art performance without manual hyperparameter tuning.
Scaling sparsemax based channel selection for speech recognition with ad-hoc microphone arrays
Because Sparsemax punishes the weights of many channels to zero harshly, we propose Scaling Sparsemax which punishes the channels mildly by setting the weights of very noisy channels to zero only.
Transformer-based end-to-end speech recognition with residual Gaussian-based self-attention
Although the performance of the proposed resGSA-Transformer is only slightly better than that of the RPSA-Transformer, it does not have to tune the window length manually.
Libri-adhoc40: A dataset collected from synchronized ad-hoc microphone arrays
We trained several multi-device speech recognition systems on both the Libri-adhoc40 dataset and a simulated dataset.
Deep NMF Topic Modeling
In this paper, we propose a deep NMF (DNMF) topic modeling framework to alleviate the aforementioned problems.
Speaker Recognition Based on Deep Learning: An Overview
Speaker recognition is a task of identifying persons from their voices.
Deep Ad-hoc Beamforming Based on Speaker Extraction for Target-Dependent Speech Separation
In this paper, we propose deep ad-hoc beamforming based on speaker extraction, which is to our knowledge the first work for target-dependent speech separation based on ad-hoc microphone arrays and deep learning.
A comparison of handcrafted, parameterized, and learnable features for speech separation
Among them, learnable features, which are trained with separation networks jointly in an end-to-end fashion, become a new trend of modern speech separation research, e. g. convolutional time domain audio separation network (Conv-Tasnet), while handcrafted and parameterized features are also shown competitive in very recent studies.
Sound
Transformer-based End-to-End Speech Recognition with Local Dense Synthesizer Attention
To reduce the computational complexity and improve the performance, we further propose local DSA (LDSA) to restrict the attention scope of DSA to a local range around the current central frame for speech recognition.
Speech enhancement aided end-to-end multi-task learning for voice activity detection
Recent studies show that speech enhancement is helpful to VAD, but the performance improvement is limited.
Partial AUC optimization based deep speaker embeddings with class-center learning for text-independent speaker verification
We also propose a class-center based training trial construction method to improve the training efficiency, which is critical for the proposed loss function to be comparable to the identification loss in performance.
Deep topic modeling by multilayer bootstrap network and lasso
Specifically, we first apply multilayer bootstrap network (MBN), which is an unsupervised deep model, to reduce the dimension of documents, and then use the low-dimensional data representations or their clustering results as the target of supervised Lasso for topic word discovery.
Multi-channel Speech Separation Using Deep Embedding Model with Multilayer Bootstrap Networks
To deal with the problem, we propose a variant of DPCL, named DPCL++, by applying a recent unsupervised deep learning method---multilayer bootstrap networks(MBN)---to further reduce the noise and small variations of the embedding vectors in an unsupervised way in the test stage, which fascinates k-means to produce a good result.
Deep Ad-hoc Beamforming
Its core idea is to reweight the estimated speech signals with a sparsity constraint when conducting adaptive beamforming, where the weights produced by a neural network are an estimation of the propagation cost from the speech source to the ad-hoc microphone array, e. g. signal-to-noise ratios, and the sparsity constraint is to filter out the microphones that are too far away from both the speech source and the majority of the ad-hoc microphone array.
Sound Audio and Speech Processing
Learning the kernel matrix by resampling
In this abstract paper, we introduce a new kernel learning method by a nonparametric density estimator.
Multilayer bootstrap network for unsupervised speaker recognition
We apply multilayer bootstrap network (MBN), a recent proposed unsupervised learning method, to unsupervised speaker recognition.
Preprint ARPPS Augmented Reality Pipeline Prospect System
This is the preprint version of our paper on ICONIP.
Unsupervised model compression for multilayer bootstrap networks
Our result suggests that the new technique integrates the effectiveness of MBN on unsupervised learning and the effectiveness and efficiency of DNN on supervised learning together for the effectiveness and efficiency of compressive MBN on unsupervised learning.
Deep Distributed Random Samplings for Supervised Learning: An Alternative to Random Forests?
In this paper, we further extend it to supervised learning incrementally.
Multilayer bootstrap networks
Geometrically, the nonparametric density estimator at each layer projects the input data space to a uniformly-distributed discrete feature space, where the similarity of two data points in the discrete feature space is measured by the number of the nearest centroids they share in common.
Learning Deep Representations By Distributed Random Samplings
In this paper, we propose an extremely simple deep model for the unsupervised nonlinear dimensionality reduction -- deep distributed random samplings, which performs like a stack of unsupervised bootstrap aggregating.
Learning Deep Representation Without Parameter Inference for Nonlinear Dimensionality Reduction
Restricted Boltzman machine, sparse coding, regularized auto-encoders, and convolutional neural networks are pioneering building blocks of deep learning.
Simple Deep Random Model Ensemble
(ii) Based on the above two views, we propose a very simple deep learning algorithm, named deep random model ensemble (DRME).
Heuristic Ternary Error-Correcting Output Codes Via Weight Optimization and Layered Clustering-Based Approach
One important classifier ensemble for multiclass classification problems is Error-Correcting Output Codes (ECOCs).
Convex Discriminative Multitask Clustering
Then, we propose two convex DMTC objectives within the framework.
