arXiv subjects

Lam Pham

Publications and source records attributed to Lam Pham.

At least 55 records · Page 3Linked to original sources

arXiv2021

Robust Deep Learning Frameworks for Acoustic Scene and Respiratory Sound Classification

This thesis focuses on dealing with the task of acoustic scene classification (ASC), and then applied the techniques developed for ASC to a real-life application of detecting respiratory disease. To deal with ASC challenges, this thesis addresses three main factors that directly affect the performance of an ASC system. Firstly, this thesis explores input features by making use of multiple spectrograms (log-mel, Gamma, and CQT) for low-level feature extraction to tackle the issue of insufficiently discriminative or descriptive input features. Next, a novel Encoder network architecture is introduced. The Encoder firstly transforms each low-level spectrogram into high-level intermediate features, or embeddings, and thus combines these high-level features to form a very distinct composite feature. The composite or combined feature is then explored in terms of classification performance, with different Decoders such as Random Forest (RF), Multilayer Perception (MLP), and Mixture of Experts (MoE). By using this Encoder-Decoder framework, it helps to reduce the computation cost of the reference process in ASC systems which make use of multiple spectrogram inputs. Since the proposed techniques applied for general ASC tasks were shown to be highly effective, this inspired an application to a specific real-life application. This was namely the 2017 Internal Conference on Biomedical Health Informatics (ICBHI) respiratory sound dataset. Building upon the proposed ASC framework, the ICBHI tasks were tackled with a deep learning framework, and the resulting system shown to be capable at detecting respiratory anomaly cycles and diseases.

cs.SD↗

arXiv2021

A Low-Compexity Deep Learning Framework For Acoustic Scene Classification

In this paper, we presents a low-complexity deep learning frameworks for acoustic scene classification (ASC). The proposed framework can be separated into three main steps: Front-end spectrogram extraction, back-end classification, and late fusion of predicted probabilities. First, we use Mel filter, Gammatone filter and Constant Q Transfrom (CQT) to transform raw audio signal into spectrograms, where both frequency and temporal features are presented. Three spectrograms are then fed into three individual back-end convolutional neural networks (CNNs), classifying into ten urban scenes. Finally, a late fusion of three predicted probabilities obtained from three CNNs is conducted to achieve the final classification result. To reduce the complexity of our proposed CNN network, we apply two model compression techniques: model restriction and decomposed convolution. Our extensive experiments, which are conducted on DCASE 2021 (IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events) Task 1A development dataset, achieve a low-complexity CNN based framework with 128 KB trainable parameters and the best classification accuracy of 66.7%, improving DCASE baseline by 19.0%

cs.SD↗

arXiv2021

Deep Learning Frameworks Applied For Audio-Visual Scene Classification

In this paper, we present deep learning frameworks for audio-visual scene classification (SC) and indicate how individual visual and audio features as well as their combination affect SC performance. Our extensive experiments, which are conducted on DCASE (IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events) Task 1B development dataset, achieve the best classification accuracy of 82.2%, 91.1%, and 93.9% with audio input only, visual input only, and both audio-visual input, respectively. The highest classification accuracy of 93.9%, obtained from an ensemble of audio-based and visual-based frameworks, shows an improvement of 16.5% compared with DCASE baseline.

cs.SD↗

arXiv2021

An Analysis of State-of-the-art Activation Functions For Supervised Deep Neural Network

This paper provides an analysis of state-of-the-art activation functions with respect to supervised classification of deep neural network. These activation functions comprise of Rectified Linear Units (ReLU), Exponential Linear Unit (ELU), Scaled Exponential Linear Unit (SELU), Gaussian Error Linear Unit (GELU), and the Inverse Square Root Linear Unit (ISRLU). To evaluate, experiments over two deep learning network architectures integrating these activation functions are conducted. The first model, basing on Multilayer Perceptron (MLP), is evaluated with MNIST dataset to perform these activation functions. Meanwhile, the second model, likely VGGish-based architecture, is applied for Acoustic Scene Classification (ASC) Task 1A in DCASE 2018 challenge, thus evaluate whether these activation functions work well in different datasets as well as different network architectures.

cs.LG↗

arXiv2021

Multi-view Audio and Music Classification

We propose in this work a multi-view learning approach for audio and music classification. Considering four typical low-level representations (i.e. different views) commonly used for audio and music recognition tasks, the proposed multi-view network consists of four subnetworks, each handling one input types. The learned embedding in the subnetworks are then concatenated to form the multi-view embedding for classification similar to a simple concatenation network. However, apart from the joint classification branch, the network also maintains four classification branches on the single-view embedding of the subnetworks. A novel method is then proposed to keep track of the learning behavior on the classification branches and adapt their weights to proportionally blend their gradients for network training. The weights are adapted in such a way that learning on a branch that is generalizing well will be encouraged whereas learning on a branch that is overfitting will be slowed down. Experiments on three different audio and music classification tasks show that the proposed multi-view network not only outperforms the single-view baselines but also is superior to the multi-view baselines based on concatenation and late fusion.

cs.SD↗

arXiv2020

Deep Learning Framework Applied for Predicting Anomaly of Respiratory Sounds

This paper proposes a robust deep learning framework used for classifying anomaly of respiratory cycles. Initially, our framework starts with front-end feature extraction step. This step aims to transform the respiratory input sound into a two-dimensional spectrogram where both spectral and temporal features are well presented. Next, an ensemble of C- DNN and Autoencoder networks is then applied to classify into four categories of respiratory anomaly cycles. In this work, we conducted experiments over 2017 Internal Conference on Biomedical Health Informatics (ICBHI) benchmark dataset. As a result, we achieve competitive performances with ICBHI average score of 0.49, ICBHI harmonic score of 0.42.

cs.LG↗

arXiv2020

Inception-Based Network and Multi-Spectrogram Ensemble Applied For Predicting Respiratory Anomalies and Lung Diseases

This paper presents an inception-based deep neural network for detecting lung diseases using respiratory sound input. Recordings of respiratory sound collected from patients are firstly transformed into spectrograms where both spectral and temporal information are well presented, referred to as front-end feature extraction. These spectrograms are then fed into the proposed network, referred to as back-end classification, for detecting whether patients suffer from lung-relevant diseases. Our experiments, conducted over the ICBHI benchmark meta-dataset of respiratory sound, achieve competitive ICBHI scores of 0.53/0.45 and 0.87/0.85 regarding respiratory anomaly and disease detection, respectively.

cs.SD↗

arXiv2020

A General Network Architecture for Sound Event Localization and Detection Using Transfer Learning and Recurrent Neural Network

Polyphonic sound event detection and localization (SELD) task is challenging because it is difficult to jointly optimize sound event detection (SED) and direction-of-arrival (DOA) estimation in the same network. We propose a general network architecture for SELD in which the SELD network comprises sub-networks that are pretrained to solve SED and DOA estimation independently, and a recurrent layer that combines the SED and DOA estimation outputs into SELD outputs. The recurrent layer does the alignment between the sound classes and DOAs of sound events while being unaware of how these outputs are produced by the upstream SED and DOA estimation algorithms. This simple network architecture is compatible with different existing SED and DOA estimation algorithms. It is highly practical since the sub-networks can be improved independently. The experimental results using the DCASE 2020 SELD dataset show that the performances of our proposed network architecture using different SED and DOA estimation algorithms and different audio formats are competitive with other state-of-the-art SELD algorithms. The source code for the proposed SELD network architecture is available at Github.

eess.AS↗

arXiv2020

Improving GANs for Speech Enhancement

Generative adversarial networks (GAN) have recently been shown to be efficient for speech enhancement. However, most, if not all, existing speech enhancement GANs (SEGAN) make use of a single generator to perform one-stage enhancement mapping. In this work, we propose to use multiple generators that are chained to perform multi-stage enhancement mapping, which gradually refines the noisy input signals in a stage-wise fashion. Furthermore, we study two scenarios: (1) the generators share their parameters and (2) the generators' parameters are independent. The former constrains the generators to learn a common mapping that is iteratively applied at all enhancement stages and results in a small model footprint. On the contrary, the latter allows the generators to flexibly learn different enhancement mappings at different stages of the network at the cost of an increased model size. We demonstrate that the proposed multi-stage enhancement approach outperforms the one-stage SEGAN baseline, where the independent generators lead to more favorable results than the tied generators. The source code is available at http://github.com/pquochuy/idsegan.

cs.LG↗

arXiv2020

On Multitask Loss Function for Audio Event Detection and Localization

Audio event localization and detection (SELD) have been commonly tackled using multitask models. Such a model usually consists of a multi-label event classification branch with sigmoid cross-entropy loss for event activity detection and a regression branch with mean squared error loss for direction-of-arrival estimation. In this work, we propose a multitask regression model, in which both (multi-label) event detection and localization are formulated as regression problems and use the mean squared error loss homogeneously for model training. We show that the common combination of heterogeneous loss functions causes the network to underfit the data whereas the homogeneous mean squared error loss leads to better convergence and performance. Experiments on the development and validation sets of the DCASE 2020 SELD task demonstrate that the proposed system also outperforms the DCASE 2020 SELD baseline across all the detection and localization metrics, reducing the overall SELD error (the combined metric) by approximately 10% absolute.

eess.AS↗

arXiv2020

CNN-MoE based framework for classification of respiratory anomalies and lung disease detection

This paper presents and explores a robust deep learning framework for auscultation analysis. This aims to classify anomalies in respiratory cycles and detect disease, from respiratory sound recordings. The framework begins with front-end feature extraction that transforms input sound into a spectrogram representation. Then, a back-end deep learning network is used to classify the spectrogram features into categories of respiratory anomaly cycles or diseases. Experiments, conducted over the ICBHI benchmark dataset of respiratory sounds, confirm three main contributions towards respiratory-sound analysis. Firstly, we carry out an extensive exploration of the effect of spectrogram type, spectral-time resolution, overlapped/non-overlapped windows, and data augmentation on final prediction accuracy. This leads us to propose a novel deep learning system, built on the proposed framework, which outperforms current state-of-the-art methods. Finally, we apply a Teacher-Student scheme to achieve a trade-off between model performance and model complexity which additionally helps to increase the potential of the proposed framework for building real-time applications.

eess.AS↗

arXiv2020

Sound Context Classification Basing on Join Learning Model and Multi-Spectrogram Features

In this paper, we present a deep learning framework applied for Acoustic Scene Classification (ASC), the task of classifying scene contexts from environmental input sounds. An ASC system generally comprises of two main steps, referred to as front-end feature extraction and back-end classification. In the first step, an extractor is used to extract low-level features from raw audio signals. Next, the discriminative features extracted are fed into and classified by a classifier, reporting accuracy results. Aim to develop a robust framework applied for ASC, we address exited issues of both the front-end and back-end components in an ASC system, thus present three main contributions: Firstly, we carry out a comprehensive analysis of spectrogram representation extracted from sound scene input, thus propose the best multi-spectrogram combinations. In terms of back-end classification, we propose a novel join learning architecture using parallel convolutional recurrent networks, which is effective to learn spatial features and temporal sequences of spectrogram input. Finally, good experimental results obtained over benchmark datasets of IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE) 2016 Task 1, 2017 Task 1, 2018 Task 1A & 1B, LITIS Rouen prove our proposed framework general and robust for ASC task.

cs.SD↗

arXiv2020

Deep Feature Embedding and Hierarchical Classification for Audio Scene Classification

In this work, we propose an approach that features deep feature embedding learning and hierarchical classification with triplet loss function for Acoustic Scene Classification (ASC). In the one hand, a deep convolutional neural network is firstly trained to learn a feature embedding from scene audio signals. Via the trained convolutional neural network, the learned embedding embeds an input into the embedding feature space and transforms it into a high-level feature vector for representation. In the other hand, in order to exploit the structure of the scene categories, the original scene classification problem is structured into a hierarchy where similar categories are grouped into meta-categories. Then, hierarchical classification is accomplished using deep neural network classifiers associated with triplet loss function. Our experiments show that the proposed system achieves good performance on both the DCASE 2018 Task 1A and 1B datasets, resulting in accuracy gains of 15.6% and 16.6% absolute over the DCASE 2018 baseline on Task 1A and 1B, respectively.

eess.AS↗

arXiv2020

Robust Acoustic Scene Classification using a Multi-Spectrogram Encoder-Decoder Framework

This article proposes an encoder-decoder network model for Acoustic Scene Classification (ASC), the task of identifying the scene of an audio recording from its acoustic signature. We make use of multiple low-level spectrogram features at the front-end, transformed into higher level features through a well-trained CNN-DNN front-end encoder. The high level features and their combination (via a trained feature combiner) are then fed into different decoder models comprising random forest regression, DNNs and a mixture of experts, for back-end classification. We report extensive experiments to evaluate the accuracy of this framework for various ASC datasets, including LITIS Rouen and IEEE AASP Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE) 2016 Task 1, 2017 Task 1, 2018 Tasks 1A & 1B and 2019 Tasks 1A & 1B. The experimental results highlight two main contributions; the first is an effective method for high-level feature extraction from multi-spectrogram input via the novel C-DNN architecture encoder network, and the second is the proposed decoder which enables the framework to achieve competitive results on various datasets. The fact that a single framework is highly competitive for several different challenges is an indicator of its robustness for performing general ASC tasks.

cs.SD↗

arXiv2020

Robust Deep Learning Framework For Predicting Respiratory Anomalies and Diseases

This paper presents a robust deep learning framework developed to detect respiratory diseases from recordings of respiratory sounds. The complete detection process firstly involves front end feature extraction where recordings are transformed into spectrograms that convey both spectral and temporal information. Then a back-end deep learning model classifies the features into classes of respiratory disease or anomaly. Experiments, conducted over the ICBHI benchmark dataset of respiratory sounds, evaluate the ability of the framework to classify sounds. Two main contributions are made in this paper. Firstly, we provide an extensive analysis of how factors such as respiratory cycle length, time resolution, and network architecture, affect final prediction accuracy. Secondly, a novel deep learning based framework is proposed for detection of respiratory diseases and shown to perform extremely well compared to state of the art methods.

cs.SD↗

arXiv2019

Spatio-Temporal Attention Pooling for Audio Scene Classification

Acoustic scenes are rich and redundant in their content. In this work, we present a spatio-temporal attention pooling layer coupled with a convolutional recurrent neural network to learn from patterns that are discriminative while suppressing those that are irrelevant for acoustic scene classification. The convolutional layers in this network learn invariant features from time-frequency input. The bidirectional recurrent layers are then able to encode the temporal dynamics of the resulting convolutional features. Afterwards, a two-dimensional attention mask is formed via the outer product of the spatial and temporal attention vectors learned from two designated attention layers to weigh and pool the recurrent output into a final feature vector for classification. The network is trained with between-class examples generated from between-class data augmentation. Experiments demonstrate that the proposed method not only outperforms a strong convolutional neural network baseline but also sets new state-of-the-art performance on the LITIS Rouen dataset.

cs.SD↗

arXiv2019

Beyond Equal-Length Snippets: How Long is Sufficient to Recognize an Audio Scene?

Due to the variability in characteristics of audio scenes, some scenes can naturally be recognized earlier than others. In this work, rather than using equal-length snippets for all scene categories, as is common in the literature, we study to which temporal extent an audio scene can be reliably recognized given state-of-the-art models. Moreover, as model fusion with deep network ensemble is prevalent in audio scene classification, we further study whether, and if so, when model fusion is necessary for this task. To achieve these goals, we employ two single-network systems relying on a convolutional neural network and a recurrent neural network for classification as well as early fusion and late fusion of these networks. Experimental results on the LITIS-Rouen dataset show that some scenes can be reliably recognized with a few seconds while other scenes require significantly longer durations. In addition, model fusion is shown to be the most beneficial when the signal length is short.

cs.SD↗

arXiv2019

Uniform Kazhdan Constants and Paradoxes of the Affine Plane

Let $G=\mathrm{SL}(2,\mathbb{Z})\ltimes\mathbb{Z}^2$ and $H=\mathrm{SL}(2,\mathbb{Z})$. We prove that the action $G\curvearrowright\mathbb{R}^2$ is uniformly non-amenable and that the quasi-regular representation of $G$ on $\ell^2(G/H)$ has a uniform spectral gap. Both results are a consequence of a uniform quantitative form of ping-pong for affine transformations, which we establish here.

math.GR↗