Andrea Spadaccini

Heart sounds quality analysis for automatic cardiac biometry applications

In this paper we propose a cardiac biometric system for human identity verification based on an automatic selection algorithm of the best subsequence of a DHS (Digital Heart Sound) signal. The quality score is based on the cepstral distance between homogeneous cardiac sounds. Performance of the algorithm proposed, expressed in terms of equal error rate, is similar to a DHS manual segmentation-based system, but offers the advantages of a fully automatic biometric application.

Human Identity Verification based on Heart Sounds: Recent Advances and Future Directions

Identity verification is an increasingly important process in our daily lives, and biometric recognition is a natural solution to the authentication problem. One of the most important research directions in the field of biometrics is the characterization of novel biometric traits that can be used in conjunction with other traits, to limit their shortcomings or to enhance their performance. The aim of this work is to introduce the reader to the usage of heart sounds for biometric recognition, describing the strengths and the weaknesses of this novel trait and analyzing in detail the methods developed so far by different research groups and their performance.

Human identity verification based on Mel frequency analysis of digital heart sounds

This paper presents new results in human identity verification via frequency analysis of cardiac sounds. More specifically, the paper proposes a pattern recognition approach based on a feature set of 13 Mel Frequency Cepstral Coefficients (MFCCs) extracted from the first (S1) and second (S2) heart sounds and a metric based on the power ratio of S1 to S2. The new algorithm yields significantly better performances with respect to the previous method based on the Chirp z-transform, guaranteeing an equal error rate (EER) below 9 %.

Supporting Undergraduate Computer Architecture Students Using a Visual MIPS64 CPU Simulator

The topics of computer architecture are always taught using an Assembly dialect as an example. The most commonly used textbooks in this field use the MIPS64 Instruction Set Architecture (ISA) to help students in learning the fundamentals of computer architecture because of its orthogonality and its suitability for real-world applications. This paper shows how to use the EduMIPS64 visual CPU Simulator as a supporting tool for teaching the standard topics covered by an undergraduate course in computer architecture. The proposed approach is first compared to other similar works in the field, then after a short description of the simulator, the paper focuses on how it can be used for teaching specific topics in an undergraduate computer architecture course. This discussion is then followed by a quantitative assessment of the suitability of the simulator by means of a survey compiled by students themselves; the results show that EduMIPS64 is suitable for the purpose for which it was built-that is, supporting the learning process of computer architecture topics.

Performance evaluation of heart sounds biometric systems on an open dataset

Recently, many systems and approaches that employ heart sounds as physiological traits for biometric recognition have been investigated. However, those systems are often tested on small, diverse and closed datasets, making it difficult to compare their performance. In this paper, we present HSCT-11, an open dataset containing data collected from 206 people that can be used for performance evaluation of heart sounds biometric systems, and we use it to benchmark two such systems. The most performing one shows an Equal Error Rate of 13.66 % on this database, a result that will be the baseline for all the future evaluations made using this dataset.

An improved biometric identification system based on heart sounds and Gaussian Mixture Models

This paper presents an evolution of a biometric identity verification system based on heart sounds. The system is built using Gaussian Mixture Models (GMMs) and uses features extracted both from the spectral domain and the time domain in order to improve the performance, measured in terms of Equal Error Rate (EER), with respect to similar systems. The best result obtained using our approach, computed over a database of 165 people, is an EER of 13,70 %, that outperforms other similar approaches.

The role of Voice Activity Detection in forensic speaker verification

This paper presents an analysis of the role of Voice Activity Detection (VAD) algorithms in forensic speaker verification systems. Those systems often have to deal with noisy phone tappings, so the activity of the separation of speech and noise, performed by VAD algorithms, is crucial. In this work we evaluate the performance of 2 widespread VAD algorithms and the corresponding performance of the speaker verification systems, using 3 kinds of additive noise (CAR, FACTORY and OFFICE) and 3 values of Signal to Noise Ratio (SNR); we then analyze the error rates showing that using a single VAD algorithm often is not the best choice in this context, but instead the VAD algorithm should be dynamically chosen according to the conditions of the audio material.

Performance Evaluation of SNR Estimation Methods in Forensic Speaker Recognition

Speech signal quality is of fundamental importance for accurate speaker identification. The reliability of a speech biometry system, in fact, is known to depend on the amount of material available, in particular on the number of vowels present in the sequence being analysed and on the quality of the signal. This paper highlights the performance of two Signal-to-Noise Ratio (SNR) estimation methods (manual and semi-automatic) usually adopted for the evaluation of speech signal quality. The results not only demonstrate the different impact of noise on the single vowels, but also show that often the SNR is over-estimated or under-estimated, and this leads respectively to the inclusion of bad quality biometric samples or the exclusion of good quality data, with a negative impact on the accuracy of the identity verification test. The paper proposes a series of issues that need to be tackled in order to develop a better procedure for the selection of biometric samples extracted from the intercepted audio signal.

Performance Evaluation of Automatic Speaker Recognition Techniques for Forensic Applications

Speaker recognition is a biometric technique employed in many different contexts, with various degrees of success. One of the most controversial usage of automatic speaker recognition systems is their employment in the forensic context [1, 2], in which the goal is to analyze speech data coming from wiretappings or ambient recordings retrieved during criminal investigation, with the purpose of recognizing if a given sentence had been uttered by a given person.

Performance evaluation of an automatic forensic speaker recognition system based on GMM

This paper presents a performance evaluation of a speech biometry system based on the statistical models GMM (Gaussian Mixture Models). In particular, the paper underlines the robustness to the degradation of various natural noises, and their impact on the system. Finally, the impact of the duration to both training and test sequences is highlighted. Results show that the noise can have the impact on the degradation of the performance (see EER values) which vary from 100 % to 300 % on the basis of the type of noise which depends on only one of two compared sequences. The duration of the sequences is a very important parameter, mostly for training phase, for which it is necessary to have at least 25 seconds long talk.