Audio Fingerprinting Research Articles

Acoustic analysis of a mechanical circulatory support.

Mechanical circulatory support technology is continually improving. However, adverse complications do occur with devastating consequences, for example, pump thrombosis that may develop in several parts of the pump system. The aim of this study was to design an experimental clot/thrombosis model to register and analyze acoustic signals from the left ventricular assist device (LVAD) HeartMate II (HMII) (Thoratec Corporation, Inc., Pleasanton, CA, USA) and detect changes in sound signals correlating to clots in the inflow, outflow, and pump housing. Using modern telecom techniques, it was possible to register and analyze the HMII pump-specific acoustic fingerprint in an experimental model of LVAD support using a mock loop. Increase in pump speed significantly (P < 0.005) changed the acoustic fingerprint at certain frequency (0–23 000 Hz) intervals (regions: R1–3 and peaks: P1,3–4). When the ball valves connected to the tubing were narrowed sequentially by ∼50% of the inner diameter (to mimic clot in the out- and inflow tubing), the frequency spectrum changed significantly (P < 0.005) in P1 and P2 and R1 when the outflow tubing was narrowed. This change was not seen to the same extent when the lumen of the ball valve connected to the inflow tube was narrowed by ∼50%. More significant (P < 0.005) acoustic changes were detected in P1 and P2 and R1 and R3, with the largest dB figs. in the lower frequency ranges in R1 and P2, when artificial clots and blood clots passed through the pump system. At higher frequencies, a significant change in dB figs. in R3 and P4 was detected when clots passed through the pump system. Acoustic monitoring of pump sounds may become a valuable tool in LVAD surveillance.

Automata and music contour functions

There is a tradition in music theory which investigates contour functions. Schützenberger's Theorem [On the definition of a family of automata. Info Control. 1961;4(2–3):245–270] connects such functions to weighted automata. Hence, it should be of music-theoretical interest to study those automata. Precisely, the minimal weighted automata computing the ascending, descending, equalizing, semitone ascending, etc. contour functions are explicitly constructed. In this article, the general combinatorial form and compactness of a musical string (finite sequences of musical entities) are discussed and the minimal automata computing the corresponding functions are displayed. Finally contour situations arising from penalty systems, classical segmentation rules, and music identification are presented.

PREDICTION AND SIMULATION OF BROADBAND PROPELLER NOISE

An empirical prediction model of broadband noise for marine propellers is proposed. The model is composed of two components. The first component is the empirical prediction of the frequency domain broadband noise based on a well-established lifting surface algorithm providing accurate prediction of the cavitation pattern on the propeller. The second component is the modulation of the noise in the time domain. The two components of the model applied in sequence generate an output yielding a realistic propeller spectrum and audio signature. The propeller noise developed in this study, with the addition of other sources such as machinery and natural noise, is used to drive the input of a submarine sonar simulator for training purposes.

Comparative study of methods for reducing dimensionality of MPEG-7 audio signature descriptors

We study how to reduce the dimensionality of the MPEG-7 audio signature descriptors in this paper. With the aid of the dimension-reduced descriptors, the comparison time for detecting copyrighted audio can be significantly reduced. The studied methods include block average, principal component analysis (PCA), Hadamard transform, Haar transform, and CDF (Cohen-Daubechies-Feauveau) 9/7 wavelet transform. For the latter four methods, we also examine whether different partition methods would affect the accuracy. The simulation results show that different reduction methods should use different partition strategies for best accuracy. In addition, we also compare the computational complexity of these methods. The experimental results show that, except the CDF 9/7 method, the rest four methods yield comparable accuracy for undistorted and MP-3 coded audio. When also considering the computational complexity, the block average method is a better choice.

MULTIMODAL INTEGRATION OF MICRO-DOPPLER SONAR AND AUDITORY SIGNALS FOR BEHAVIOR CLASSIFICATION WITH CONVOLUTIONAL NETWORKS

The ability to recognize the behavior of individuals is of great interest in the general field of safety (e.g. building security, crowd control, transport analysis, independent living for the elderly). Here we report a new real-time acoustic system for human action and behavior recognition that integrates passive audio and active micro-Doppler sonar signatures over multiple time scales. The system architecture is based on a six-layer convolutional neural network, trained and evaluated using a dataset of 10 subjects performing seven different behaviors. Probabilistic combination of system output through time for each modality separately yields 94% (passive audio) and 91% (micro-Doppler sonar) correct behavior classification; probabilistic multimodal integration increases classification performance to 98%. This study supports the efficacy of micro-Doppler sonar systems in characterizing human actions, which can then be efficiently classified using ConvNets. It also demonstrates that the integration of multiple sources of acoustic information can significantly improve the system's performance.

An efficient audio fingerprint search algorithm for music retrieval

The conventional audio fingerprinting system by Haitsma uses a lookup table to identify the candidate songs in the database, which contains the sub-fingerprints of songs, and searches the candidates to find a song whose bit error rate is the lowest. However, this approach has a drawback that the number of database accesses increases dramatically, especially when the database contains a large number of songs or when a matching sub-fingerprint is not found in the lookup table due to a heavily degraded input signal. In this paper, a novel search method is proposed to overcome these difficulties. The proposed method partitions each song found from the lookup table into blocks, assigns a weight to each block, and uses the weight as a search priority to speed up the search process while reducing the number of database accesses. Various results from our experiment show the significant improvement in search speed while maintaining the search accuracy comparable to the conventional method.

Low-order auditory Zernike moment: a novel approach for robust music identification in the compressed domain

Audio identification via fingerprint has been an active research field for years. However, most previously reported methods work on the raw audio format in spite of the fact that nowadays compressed format audio, especially MP3 music, has grown into the dominant way to store music on personal computers and/or transmit it over the Internet. It will be interesting if a compressed unknown audio fragment could be directly recognized from the database without decompressing it into the wave format at first. So far, very few algorithms run directly on the compressed domain for music information retrieval, and most of them take advantage of the modified discrete cosine transform coefficients or derived cepstrum and energy type of features. As a first attempt, we propose in this paper utilizing compressed domain auditory Zernike moment adapted from image processing techniques as the key feature to devise a novel robust audio identification algorithm. Such fingerprint exhibits strong robustness, due to its statistically stable nature, against various audio signal distortions such as recompression, noise contamination, echo adding, equalization, band-pass filtering, pitch shifting, and slight time scale modification. Experimental results show that in a music database which is composed of 21,185 MP3 songs, a 10-s long music segment is able to identify its original near-duplicate recording, with average top-5 hit rate up to 90% or above even under severe audio signal distortions.

English

Vehicles produce sound signals with varying temporal and spectral properties under different working conditions. These sounds are indicative of the condition of the engine. Fault diagnosis is a significantly difficult task in geographically remote places where expertise is scarce. Automated fault diagnosis can assist riders to assess the health condition of their vehicles. This paper presents a method for fault detection and location in motorcycles based on the chain code of the pseudospectra and Mel-frequency cepstral coefficient (MFCC) features of acoustic signals. The work comprises two stages: fault detection and fault location. The fault detection stage uses the chain code of the pseudospectrum as a feature vector. If the motorcycle is identified as faulty, the MFCCs of the same sample are computed and used as features for fault location. Both stages employ dynamic time warping for the classification of faults. Five types of faults in motorcycles are considered in this work. Observed classification rates are over 90% for the fault detection stage and over 94% for the fault location stage. The work identifies other interesting applications in the development of acoustic fingerprints for fault diagnosis of machinery, tuning of musical instruments, medical diagnosis, etc.

Collusion-Resistant Audio Fingerprinting System in the Modulated Complex Lapped Transform Domain

Collusion-resistant fingerprinting paradigm seems to be a practical solution to the piracy problem as it allows media owners to detect any unauthorized copy and trace it back to the dishonest users. Despite the billionaire losses in the music industry, most of the collusion-resistant fingerprinting systems are devoted to digital images and very few to audio signals. In this paper, state-of-the-art collusion-resistant fingerprinting ideas are extended to audio signals and the corresponding parameters and operation conditions are proposed. Moreover, in order to carry out fingerprint detection using just a fraction of the pirate audio clip, block-based embedding and its corresponding detector is proposed. Extensive simulations show the robustness of the proposed system against average collusion attack. Moreover, by using an efficient Fast Fourier Transform core and standard computer machines it is shown that the proposed system is suitable for real-world scenarios.

Two-pass search strategy using accumulated band energy histogram for HMM-based identification of perceptually identical music

In this article, we present an efficient two-pass search strategy for the implementation of a Hidden Markov Model (HMM)-based music identification system. In our previous work, we demonstrated a single-pass HMM-based music identification system, considering its application to music copyright protection. This conventional system showed very robust performance to signal-level variations between perceptually identical music files. However, it requires heavy computation for search. In the proposed two-pass search system, the conventional single-pass search is extended to two-pass. In pass 1 of the proposed method, a queried music produces an accumulated band energy histogram which is a set of normalized sums of band energies for each frequency bin. This histogram is compared to all of the histograms for the registered music files. The system generates a list of small number of most probable music files among the all of the registered music files. In pass 2, HMM-based search is applied only to candidate music files selected in pass 1. Using the proposed two-pass strategy, we successfully implemented a HMM-based music identification system, which maintains the same level robustness to signal level variations between perceptually identical music files but also produces the identification result very quickly. © 2013 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 23, 127–132, 2013

A Review of Audio Fingerprinting and Comparison of Algorithms

An audio finger print is a small set of features that uniquely identifies a song. An audio fingerprint can be used for broadcast monitoring, audience measurement, meta-data collection. The general framework for building an audio fingerprint includes a front- end and a finger print modeling block. This paper details various uses and properties of an audio fingerprint and also the various stages included in the front end. Two algorithms namely - PRH and MLH have been discussed.

오디오 핑거프린트의 비트에러율을 이용한 자동 음악 요약 기법 및 시스템

본 논문은 음악의 코러스(chorus) 구간을 자동으로 추출하는 기법 및 시스템에 대하여 다루었다. 코러스 구간을 자동으로 추출하는 음악 요약 기술은 방대한 음악 데이터베이스에서 특정 음악 검색을 용이하게 할 수 있으며, 온라인 스트리밍 서비스에서 샘플 음악을 생성할 때 사용될 수 있다. 이를 구현하기 위해, 기존의 알고리즘들은 2차원 유사도 행렬, 확률모델, 신경망모델, 템포 특징 벡터, 클러스터링 기법 등을 적절히 활용하여 개발되었다. 본 논문에서는 음악의 오디오 핑거프린트를 추출한 후 곡 내의 오디오 핑거프린트 구간 쌍의 비트에러율을 통해 음악 요약을 추출한다. 다만, 음악 검색 솔루션에서 사용된 오디오 핑거프린트가 데이터베이스에 이미 존재할 경우에는 이를 바로 로딩한 후 비트에러율을 계산하여 음악 요약을 추출할 수 있다. 이런 방법은 이미 만들어진 데이터베이스를 변형 없이 그대로 사용할 수 있음으로써 음악 데이터베이스를 활용한 다양한 알고리즘과 솔루션의 가능성을 보여주었다. 또한, 음악의 코러스를 추출하는데 있어서 기존 방식보다 매우 뛰어난 성능을 보임을 알 수 있었다. In this paper, we present an effective method and a system for the music summarization which automatically extract the chorus portion of a piece of music. A music summary technology is very useful for browsing a song or generating a sample music for an online music service. To develop the solution, conventional automatic music summarization methods use a 2-dimensional similarity matrix, statistical models, or clustering techniques. But our proposed method extracts the music summary by calculating BER(Bit Error Rate) between audio fingerprint blocks which are extracted from a song. But we could directly use an enormous audio fingerprint database which was already saved for a music retrieval solution. This shows the possibility of developing a various of new algorithms and solutions using the audio fingerprint database. In addition, experiments show that the proposed method captures the chorus of a song more effectively than a conventional method.

A Filtering Method for Audio Fingerprint Based on Multiple Measurements

A Filtering Method for Audio Fingerprint Based on Multiple Measurements

A Video Content Classification Algorithm Applying to Human Action Recognition

A new classification algorithm of human action recognition of video content is suggested in our paper. It analyzes the variation of the content of video scenes or human action from video bi-modal features, in order to cognize content efficiently and precisely. This scheme is based on the pattern analysis of spatio-temporal slices and audio signature feature extracted from the video files. The Spatial-temporal Variation Histogram feature is firstly defined in our paper. It is applied to describe spatio-temporal variation analysis of human action or video scenes. The audio signature is also applied to identify the audio content by extracting unique signatures from a part of audio signal. The experiments show excellent performance of classification on the KTH dataset.DOI: http://dx.doi.org/10.5755/j01.eee.19.4.4056

Macro Segmentation and Content Analysis of TV Broadcast Stream

This study addresses a non-supervised approach to extract TV programs via repetition based detection of the Inter-Programs (IPs) and audio based segmentation and classification algorithm to analyze the massive raw TV stream. Acoustic and visual information are both adopted for IPs detection so as to avoid missing true-positive. Novel audio fingerprints scheme and shot based indexing algorithm are introduced to guarantee the efficient and superior detection performance. After the TV programs are further segmented into clips, Gaussian Mixture Models (GMMs) are used to classify the clips into three types, namely, pure speech, non-pure speech, and non-speech. Experiments on a test dataset composed of more than 500 hours content-unknown TV streams show that the F-measure of the programs extraction and content analysis achieve 0.986 and 0.887 respectively. The experiments also demonstrate that the proposed algorithm for detecting repeated IPs outperforms the state-of-art approach.

Efficient Audio Fingerprint Application Verification Using the Adapted Computational Geometry Algorithm

An earlier work of the authors introduced an adapted version of the Computational Geometry Algorithm (CGA) designed to analyse an audio stream and produce a unique coding-independent fingerprint. As the adaptability and the induced calculation load of the proposed algorithm form a key characteristic for multiple applications, our current investigation aims to measure its performance and stability in dynamic, real-time applications, i.e., in large audio library indexing and dynamic audio recognition. In addition, we investigate the fact that context similarity is also evident across fingerprints; hence a number of comparisons are used to explore the possible uses of this highly desirable algorithmic feature.

Automatic highlights extraction for drama video using music emotion and human face features

This paper describes a novel system that uses music emotion and human face as features for automatic highlights extraction for drama video. These high-level audiovisual features are used because music evokes emotion response from the viewer and characters express emotion on their faces. In addition, a novel scheme is developed to improve the accuracy of music emotion recognition in drama video. Specifically, emotion recognition is performed not on the input audio signal but on the noise-free music available from the album of the incidental music, with the presence of incidental music detected by an audio fingerprint technique. Besides the conventional subjective evaluation, we propose a new metric for quantitative performance evaluation of highlights extraction. Experiments conducted over four different types of drama videos demonstrate that the proposed system significantly outperforms baseline ones in terms of both subjective and objective measures.

Index Compression for Audio Fingerprinting Systems Based on Compressed Suffix Array

 Abstract—As one of most popular technologies, audio fingerprinting has recently attracted much attention in music retrieval systems. In music retrieval methods based on audio fingerprints, a large database is required in order to compare the fingerprints extracted from the query. In other words, the efficient search method has to be developed. In this paper, we propose a method for index compression using a compressed suffix array. Taking advantage of the fact that the repetitive characters occur frequently in higher bits of the sorted audio fingerprint data, the proposed method compresses the index by encoding the 8-bit data sequences by Run Length Encoding. Vertical Code is also used to compress the array, wherein the positions of the sorted data are stored. Four sets of music databases are used in experiments to evaluate the effectiveness of the proposed method. The experimental results show that the proposed method, compared with the conventional method, only needs 30% of the space of an audio fingerprints database for a music database consisting of 8000 songs, and around 80% of the index space for a database of 1000 songs. Moreover, the entire space cost is reduced to around 60%, compared with the method based on the suffix array. using a suffix array (6) has also been proposed. In the method based on the suffix array, the space cost increases in proportion to the growing music database. In this paper, we proposed a method to reduce the space cost by compressing the index of the database. The paper is organized as follows: Section II outlines music retrieval based on audio fingerprints. We review a fast Hamming space search method (7) based on a suffix array in Section III, and propose a space-saving method based on a compressed suffix array in Section IV. We evaluate the proposed method in Section V. Finally, the conclusions and future work are given in Section VI.

Audio Watermarking Based on Vector Quantization Index Modulation Using Audio Fingerprint

SUMMARY Development of high-quality digital audio watermarking with robustness to data compression and watermark attacks such as cropping is an important research topic. Recently, an effective watermarking method using vector quantization index modulation (VQIM) was proposed to improve the watermark performance. In this paper, a new auditory-band MDCT-based VQIM-type watermarking using audio fingerprint (AFP) is proposed to enhance the performance. The parameter values of VQIM are essential for audio quality and robustness, which are determined as individual codebook parameters. The parameters are estimated using AFP in the watermark detection process. In simulations, the effectiveness of the method was examined to show its robustness to watermark attacks. The audio quality was evaluated based on PEAQ.

Audio Fingerprint Binarization by Minimizing Hinge-Loss Function

This paper proposes a robust binary audio fingerprinting method by minimizing hinge-loss function. In the proposed method, the type of fingerprints is binary, which is conducive in reducing the size of fingerprint DB. In general, the binarization of features for fingerprinting deteriorates the performance of fingerprinting system, such as robustness and discriminability. Thus it is necessary to minimize such performance loss. Since the similarity between two audio clips is represented by a hinge-like function, we propose a method to derive a binary fingerprinting by minimizing a hinge-loss function. The derived hinge-loss function is minimized by using the minimal loss hashing. Experiments over thousands of songs demonstrate that the identification performance of binary fingerprinting can be improved by minimizing the proposed hinge loss function.