Filters








1,019 Hits in 4.6 sec

The Power Cepstrum Calculation with Convolutional Neural Networks

Mario Alejandro García, Eduardo Atilio Destéfanis
2019 Journal of Computer Science and Technology  
A model of neural network with convolutional layers that calculates the power cepstrum of the input signal is proposed.  ...  The main conclusions indicate, on the one hand, that it is possible to calculate the power cepstrum with a neural network; on the other hand, that it is possible to use these networks as the initial layers  ...  Acknowledgements We gratefully acknowledge the support of NVIDIA Corporation through the NVIDIA GPU Grant Program.  ... 
doi:10.24215/16666038.19.e13 fatcat:dj7tas5b6fg6tba2zgnirqk4le

An improved convolutional neural network for convenient rail damage detection

Zhongzhou Zhang, Xinhao Che, Yan Song
2022 Frontiers in Energy Research  
Finally, the performance of the three convolutional neural networks (CNN) in rail damage detection is evaluated and compared.  ...  The key information of the grayscale maps is extracted using neural networks.  ...  Acknowledgments We thank the Research Project on Multidimensional Data Characterization Methods and Structured Organization Mechanisms of Production Factors of Shenyang Institute of Automation Chinese  ... 
doi:10.3389/fenrg.2022.1007188 fatcat:kwry2jovebbthfsbdu6obkxrqi

Convolutional neural networks for passive monitoring of a shallow water environment using a single sensor

Eric L. Ferguson, Rishi Ramakrishnan, Stefan B. Williams, Craig T. Jin
2017 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
This paper proposes the use of convolutional neural networks (CNNs) for the joint detection and ranging of broadband acoustic noise sources such as marine vessels in conjunction with a data augmentation  ...  It is shown that CNNs operating on cepstrum data are able to detect the presence and estimate the range of transiting vessels at greater distances than the conventional method.  ...  For the current approach, the power cepstrum (referred to in this paper as the cepstrum) is used and is derived from the power spectrum of a recorded signal.  ... 
doi:10.1109/icassp.2017.7952638 dblp:conf/icassp/FergusonRWJ17 fatcat:ddny7mrzdjdyzb7ebzuidlqdum

Using Deep Learning Techniques and Inferential Speech Statistics for AI Synthesised Speech Recognition [article]

Arun Kumar Singh
2021 arXiv   pre-print
Here, we propose a model based on Convolutional Neural Network (CNN) and Bidirectional Recurrent Neural Network (BiRNN) that helps to achieve both the aforementioned objectives.  ...  The model outperforms the state-of-the-art approaches by classifying the AI synthesized audio from real human speech with an error rate of 1.9% and detecting the underlying architecture with an accuracy  ...  Convolutional Neural Networks (CNN) The CNNs are the neural networks that uses linear matrix operation called Convolution to extract meaningful features from the image.  ... 
arXiv:2107.11412v1 fatcat:q3sgzfxclzgxzf32menjpovrj4

Distributed Sound Transmission and Smart City Planning Management Based on Convolutional Neural Network

Yingchuan Ma, Yue Yu, Zhaoming Ma, Mohammad Farukh Hashmi
2022 Wireless Communications and Mobile Computing  
In this article, we provide a detection method based on a convolutional neural network algorithm. Preamble detection is very important in underwater acoustic communication.  ...  The prerequisite for specifying the joint position of the distributed acoustic multiarray is the mutual calibration calculation of each array.  ...  Introduction This paper presents Convolutional Neural Networks and Taigman's Entire Neural Network, combining a local music classification setup with an overall distribution.  ... 
doi:10.1155/2022/5731473 fatcat:f6pf5fimk5fq7lfc37jmcsehnq

Detection of asphyxia in infants using deep learning Convolutional Neural Network (CNN) trained on Mel Frequency Cepstrum Coefficient (MFCC) features extracted from cry sounds

A. Zabidi, I.M. Yassin, H.A. Hassan, N. Ismail, M.M.A.M. Hamzah, Z.I. Rizman, H.Z. Abidin
2018 Journal of Fundamental and Applied Sciences  
In this paper, we prove that Mel Frequency Cepstrum Coefficient (MFCC) feature generates from audio signal of infant cry could be used as input feature for the Convolution Neural Network (CNN)  ...  Deep Learning Neural Network (DLNN), is a new branch of machine learning with the ability for complex feature representatio Although it was mainly suited for image feature (since it was inspired by object  ...  There are several types of DLNNs namely Deep Belief Neural Network (DBNN), Stacked Auto-Encoders (SAE) and Convolution Neural Network (CNN).  ... 
doi:10.4314/jfas.v9i3s.59 fatcat:rts4put2jvgbtnqoyms2c6sa44

Sound Source Localization in a Multipath Environment Using Convolutional Neural Networks [article]

Eric L. Ferguson, Stefan B. Williams, Craig T. Jin
2017 arXiv   pre-print
This paper proposes the use of convolutional neural networks (CNNs) for the localization of sources of broadband acoustic radiated noise (such as motor vessels) in shallow water multipath environments.  ...  The propagation of sound in a shallow water environment is characterized by boundary reflections from the sea surface and sea floor.  ...  For the current approach, the power cepstrum is used and is derived from the power spectrum of a recorded signal.  ... 
arXiv:1710.10948v1 fatcat:ocmdyj4y4vamxflhd2uhy3puse

Wavelet and Neural Network Method for Speech Enhancement Objective Evaluation

K. Daqrouq, M. Alfaouri, T. Abu Hilal, J. Daqrouq, S. El-Hajjar
2010 The International Conference on Electrical Engineering  
Wavelet Neural Network Evaluation method WNNEM is proposed as a powerful tool for enhanced speech signal evaluation.  ...  The advantage of this method is the evaluation of different band passes of frequency based on wavelet transform by neural network, which is very powerful classification tool.  ...  The advantage of this method is the evaluation of different band passes of frequency based on wavelet transform by neural network which is very powerful classification tool.  ... 
doi:10.21608/iceeng.2010.33258 fatcat:v5jq5sfr6bhnjcsmsyqal4ydlq

Acoustic Scene Analysis Using Partially Connected Microphones Based on Graph Cepstrum [article]

Keisuke Imoto
2018 arXiv   pre-print
Our experimental results indicate that the proposed graph-based cepstrum effectively extracts spatial information with consideration of the microphone connections.  ...  Specifically, in the proposed graph-based cepstrum, the logarithm of the amplitude in a multichannel observation is converted to a feature vector by an inverse graph Fourier transform, which can consider  ...  ACKNOWLEDGMENTS Part of this work was supported by the Support Center for Advanced Telecommunications Technology Research, Foundation.  ... 
arXiv:1805.11782v2 fatcat:te7stn5pvnbxla4ax5ewfw6sju

Damage identification based on response-only measurements using cepstrum analysis and artificial neural networks

Ulrike Dackermann, Wade A Smith, Robert B Randall
2014 Structural Health Monitoring  
This paper presents a response-only structural health monitoring (SHM) technique that utilises cepstrum analysis and artificial neural networks (ANNs) for the identification of damage in civil engineering  ...  In the numerical investigation, the same input is applied to the FE model, but the obtained responses are polluted with different levels of white Gaussian noise to better replicate real-life conditions  ...  Funding The section on cepstrum-based operational modal analysis was supported in part by the Australian Research Council, through Linkage Project [LP110200738].  ... 
doi:10.1177/1475921714542890 fatcat:2waqaag2pbaefc5bcg22jazivu

DCASE 2018 Challenge: Solution for Task 5 [article]

Jeremy Chew, Yingxiang Sun, Lahiru Jayasinghe, Chau Yuen
2018 arXiv   pre-print
The proposed system consists of three different models, based on convolutional neural network and long short memory recurrent neural network.  ...  With extracted features such as spectrogram and mel-frequency cepstrum coefficients from different channels, the proposed system can classify different domestic activities effectively.  ...  The baseline system employed a single classifier model that takes a single channel as input. Its is Neural Network architecture with two convolutional layers and one dense layer.  ... 
arXiv:1812.04618v1 fatcat:ror5dnxsgrd6bbsafn33tw63aa

An application of Artificial Intelligence in emotion detection by speech only

Arvind Sharma, RJIT Tekanpur, R K Gupta, MITS Gwalior
2021 Ymer  
The robust indistinguishable features are then applied with an efficient and fast deep learning approach Neural Structured Learning (NSL) for emotion training and recognition.  ...  For a person independent emotion recognition system, audio data is used as input to the system from which, Mel Frequency Cepstral Coefficients (MFCC) are calculated as features.  ...  Later, Convolutional Neural Networks (CNN) became very popular be-cause of its improved discriminative power compared to DBN [11] .  ... 
doi:10.37896/ymer20.10/4 fatcat:tyctksfpbvebfo3nf7yejgcfpy

Physiological-Physical Feature Fusion for Automatic Voice Spoofing Detection [article]

Junxiao Xue, Hao Zhou, Yabo Wang
2021 arXiv   pre-print
This method involves feature extraction, a densely connected convolutional neural network with squeeze and excitation block (SE-DenseNet), multi-scale residual neural network with squeeze and excitation  ...  We first pre-trained a convolutional neural network using the speaker's voice and face in the video as surveillance signals. It can extract physiological features from speech.  ...  Compared with the convolutional neural network, the recurrent neural network has the advantages of accepting the sequence data of variable length as input and has the memory function.  ... 
arXiv:2109.00913v1 fatcat:a2y477ankzhvllzfluea4x6bhe

Pre-Configured Deep Convolutional Neural Networks with Various Time-Frequency Representations for Biometrics from ECG Signals

Yeong-Hyeon Byeon, Keun-Chang Kwak
2019 Applied Sciences  
We evaluated electrocardiogram (ECG) biometrics using pre-configured models of convolutional neural networks (CNNs) with various time-frequency representations.  ...  Studies have applied time-frequency representations of the 1D ECG signals to 2D CNNs using transforms like MFCC (mel frequency cepstrum coefficient), spectrogram, log spectrogram, mel spectrogram, and  ...  Conflicts of Interest: The authors declare no conflicts of interest.  ... 
doi:10.3390/app9224810 fatcat:3aarfe4kfbg2dh6l46b25ao6he

Octave-band Filtering for Convolutional Neural Network-based Diagnostics for Rotating Machinery

Namkyoung Lee, Michael Azarian, Michael Pecht
2020 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
Research into feature extraction methods for convolutional neural network (CNN)-based diagnostics for rotating machinery remains in a developmental stage.  ...  This paper introduces octave-band filtering as a feature extraction method for preprocessing a spectrogram prior to use with CNN.  ...  Figure 3 . 3 An overview of convolutional neural network model for rotating machinery diagnosis. Figure 4 . 4 A convolutional neural network structure for rotating machinery diagnosis.  ... 
doi:10.36001/phmconf.2020.v12i1.1132 fatcat:eb3hak67tncvtczjubhsbb6tky
« Previous Showing results 1 — 15 out of 1,019 results