Flexible Biometric Online Speaker-Verification System Implemented on FPGA Using Vector Floating-Point Units

Canto-Navarro, Enrique; López-García, Mariano; Ramos-Lara, Rafael; Sánchez-Reillo, Raúl

doi:10.1109/tvlsi.2014.2377578

Cited by 7 publications

(4 citation statements)

References 27 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The execution time of GMM trained on MFCCs is 0.8ms at 48MHz frequency for a speech set of length 20 [257]. Moreover, the execution time of SVM and MFCCs is 4.6ms at [258]. Another group of researchers [259], developed a SID framework based on MFCCs and SVM and concluded an execution time of 9.10ms per frame.…”

Section: ) Speaker Identification (Sid) System For Forensicsmentioning

confidence: 99%

A Survey on Artificial Intelligence-Based Acoustic Source Identification

Zaheer,

Ahmad,

Habibi

et al. 2023

IEEE Access

View full text Add to dashboard Cite

The concept of Acoustic Source Identification (ASI), which refers to the process of identifying noise sources has attracted increasing attention in recent years. The ASI technology can be used for surveillance, monitoring, and maintenance applications in a wide range of sectors, such as defence, manufacturing, healthcare, and agriculture. Acoustic signature analysis and pattern recognition remain the core technologies for noise source identification. Manual identification of acoustic signatures, however, has become increasingly challenging as dataset sizes grow. As a result, the use of Artificial Intelligence (AI) techniques for identifying noise sources has become increasingly relevant and useful. In this paper, we provide a comprehensive review of AI-based acoustic source identification techniques. We analyze the strengths and weaknesses of AI-based ASI processes and associated methods proposed by researchers in the literature. Additionally, we did a detailed survey of ASI applications in machinery, underwater applications, environment/event source recognition, healthcare, and other fields. We also highlight relevant research directions.

show abstract

Section: ) Speaker Identification (Sid) System For Forensicsmentioning

confidence: 99%

A Survey on Artificial Intelligence-Based Acoustic Source Identification

Zaheer,

Ahmad,

Habibi

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Therefore, the proposed model has attained an EER rate of 3.4%. Navarro et al (2015) has presented another interesting work on voice verification system. Here, the system hardware is enhanced with a vector floatingpoint unit in the microprocessor that increased the vector floating-point operation rapidity in the speaker verification algorithms.…”

Section: Voice Biometric Recognitionmentioning

confidence: 99%

“…The proposed scheme is evaluated under three diverse datasets and has diminished computational complexity and simulation time But in some cases, it lacks in security because of high noise signal Gałka et al (2014) Embedded scheme Here, the experimental works are done to validate the system reliability and function rate However, the voice recognition model using micro processer paradigm is quite complicated task because of its memory size and processor components Navarro et al (2015) Microprocessor-MicroBlaze in FPGA The scheme required less period for voice recognition…”

Section: Authentication Using Keystroke Dynamicsmentioning

confidence: 99%

A broad review on non-intrusive active user authentication in biometrics

Thomas¹,

Mathew

2021

J Ambient Intell Human Comput

View full text Add to dashboard Cite

Authentication is the process of keeping the user's personal information as confidential in digital applications. Moreover, the user authentication process in the digital platform is employed to verify the own users by some authentication methods like biometrics, voice recognition, and so on. Traditionally, a one-time login based credential verification method was utilized for user authentication. Recently, several new approaches were proposed to enhance the user authentication framework but those approaches have been found inconsistent during the authentication execution process. Hence, the main motive of this review article is to analyze the advantage and disadvantages of authentication systems such as voice recognition, keystroke, and mouse dynamics. These authentication models are evaluated in a continuous non-user authentication environment and their results have been presented in way of tabular and graphical representation. Also, the common merits and demerits of the discussed authentication systems are broadly explained discussion section. Henceforth, this study will help the researchers to adopt the best suitable method at each stage to build an authentication framework for non-intrusive active authentication.

show abstract

“…The execution time is 4.6 ms at 50Hz main frequency [23]. Enrique developed an accelerator on the basis of MFCC and SVM, with a frame execution time dimension of 9.10ms [24]. This kind of algorithm can obtain favorable results on pure datasets with the small data scale but is unsuitable in a real dataset with large data scale.…”

Section: Related Workmentioning

confidence: 99%

A Simplified Speaker Recognition System Based on FPGA Platform

Jiang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Speaker recognition is a crucial bio-identification technology, which is extensively used in our daily life. With the development of deep learning, convolutional neural networks (CNNs) are applied to speaker recognition tasks given their excellent performance. However, in real life, speaker recognition systems are frequently deployed on end-devices. Therefore, while obtaining recognition accuracy, the model of speaker recognition is expected to be as simple as possible. Inspired by 1-max pooling CNN and Gaussian mixture model-universal background model (GMM-UBM), this study proposes a one dimension convolutional neural networks (1D CNN) on the basis of original 2D CNN. The proposed model reduces the computational complexity of ResNet20 by 64% and the amount of parameters by 53%. In comparison with the original ResNet20 models, the recognition accuracy will be reduced by about one percent on the 15s data set. Then, on the basis of the 1D CNN, we propose a pyramid layer-folding pipeline structure and implement it on the Xilinx VC709 platform. According to the time-dimension partition, the proposed pyramid pipeline structure can process speech data of various lengths. Moreover, our accelerator is 5.1× faster on 3s dataset and 6.8× quicker on 15s dataset than those of the CPU platform.INDEX TERMS Speaker recognition,1D convolution neural networks, pyramid pipeline, folding pipeline, FPGA.

show abstract

Flexible Biometric Online Speaker-Verification System Implemented on FPGA Using Vector Floating-Point Units

Cited by 7 publications

References 27 publications

A Survey on Artificial Intelligence-Based Acoustic Source Identification

A Survey on Artificial Intelligence-Based Acoustic Source Identification

A broad review on non-intrusive active user authentication in biometrics

A Simplified Speaker Recognition System Based on FPGA Platform

Contact Info

Product

Resources

About