Much work is focused upon music genre recognition (MGR) from audio recordings, symbolic data, and other modalities. While reviews have been written of some of this work before, no survey has been made of the approaches to evaluating approaches to MGR. This paper compiles a bibliography of work in MGR, and analyzes three aspects of evaluation: experimental designs, datasets, and figures of merit.

/pdf/a-survey-of-evaluation-in-music-genre-recognition-29v3l1icya.pdf

A survey of evaluation in music genre recognition

An increase in the accuracy of identification of Activities of Daily Living (ADL) is very important for different goals of Enhanced Living Environments and for Ambient Assisted Living (AAL) tasks. This increase may be achieved through identification of the surrounding environment. Although this is usually used to identify the location, ADL recognition can be improved with the identification of the sound in that particular environment. This paper reviews audio fingerprinting techniques that can be used with the acoustic data acquired from mobile devices. A comprehensive literature search was conducted in order to identify relevant English language works aimed at the identification of the environment of ADLs using data acquired with mobile devices, published between 2002 and 2017. In total, 40 studies were analyzed and selected from 115 citations. The results highlight several audio fingerprinting techniques, including Modified discrete cosine transform (MDCT), Mel-frequency cepstrum coefficients (MFCC), Principal Component Analysis (PCA), Fast Fourier Transform (FFT), Gaussian mixture models (GMM), likelihood estimation, logarithmic moduled complex lapped transform (LMCLT), support vector machine (SVM), constant Q transform (CQT), symmetric pairwise boosting (SPB), Philips robust hash (PRH), linear discriminant analysis (LDA) and discrete cosine transform (DCT).

https://rua.ua.es/dspace/bitstream/10045/72592/1/2018_Pires_etal_Sensors.pdf

Recognition of Activities of Daily Living Based on Environmental Analyses Using Audio Fingerprinting Techniques: A Systematic Review.

Audio fingerprinting, also named as audio hashing, has been well-known as a powerful technique to perform audio identification and synchronization. It basically involves two major steps: fingerprint (voice pattern) design and matching search. While the first step concerns the derivation of a robust and compact audio signature, the second step usually requires knowledge about database and quick-search algorithms. Though this technique offers a wide range of real-world applications, to the best of the authors’ knowledge, a comprehensive survey of existing algorithms appeared more than eight years ago. Thus, in this paper, we present a more up-to-date review and, for emphasizing on the audio signal processing aspect, we focus our state-of-the-art survey on the fingerprint design step for which various audio features and their tractable statistical models are discussed. Keywords–Voice pattern; audio identification and synchronization; spectral features; statistical models.

A Review of Audio Features and Statistical Models Exploited for Voice Pattern Design

Perceptual audio hashing algorithm based on Zernike moment and maximum-likelihood watermark detection

Combined with the linear prediction-minimum mean squared error (LP-MMSE), an efficient perceptual hashing algorithm based on improved spectral entropy for speech authentication was proposed in this paper. The linear prediction analysis is conducted on speech signal after preprocessing, framing and adding windows, and obtained the minimum mean squared error coefficient matrix. And then, the spectral entropy parameter matrix of each frame is calculated by using improved spectral entropy method. And the final binary perceptual hashing sequence is generated based on the above two matrices, and the speech authentication is completed. Comparing the experimental results of combining the Teager energy operator (TEO) with the linear predictive coefficients (LPC), LP-MMSE and line spectrum pair (LSP) coefficient respectively, it can be seen that the proposed algorithm had a good compromise between robustness, discrimination and authentication efficiency, and the proposed algorithm can meet the requirement of real-time speech authentication in speech communication. Experimental results show that the proposed algorithm was better than other existing methods in compactness.

An efficient perceptual hashing based on improved spectral entropy for speech authentication

Audio fingerprint is an effective representation of an audio signal using low-level features and can be used to identify unlabeled audio based on its content. In this paper, we introduce a robust audio feature, local energy centroid (LEC), which can represent the energy conglomeration degree of the relative small region in the spectrum. Our audio fingerprint is generated based on the LEC feature which is conducive to enhance the robustness of system. In audio retrieval processing, an improved scoring strategy is proposed to resist the linear speed change. Experimental results show that the new fingerprinting system is quite robust in the present of noise and the proposed method can achieve satisfying recognition accuracy.

Audio fingerprinting based on local energy centroid

Audio fingerprint is also named audio hashing, which is a content-based identification technique that can be used to label an unknown audio recording. Here we introduce an audio fingerprinting algorithm based on spectral energy structure and NMF(non-negative matrix factorization) algorithm. Firstly, a matrix of audio features is generated based on the spectral subband energy of audio signal. Then NMF is performed on this matrix to capture local characteristics of audio signal, which is expected to reduce the possibility of Declassification. Experimental results suggest that the proposed scheme can achieve good recognition accuracy and can be used in applications such as broadcast monitoring and audio excerpt identification by a mobile phone.

Audio fingerprinting based on spectral energy structure and NMF

Audio fingerprint is a compact unique content-based digital signature of an audio signal. It's an interesting technique that can be used to identify unknown audio clips. Generally, it mainly consists of two parts, i.e. fingerprint extracting from audio signals and fingerprint matching against those stored in a fingerprint database that has been set up beforehand. With the rapid growth in the quantity of audio files, the probability of collision of different audio signals become relatively high and it has become very challenging to retrieve an audio recording in real-time from the ever-growing huge database. In this letter, we introduce a reliable audio fingerprinting system, which extracts audio fingerprints from an audio signal based on its spectral energy structure. Preliminary experimental results suggest that this fingerprinting system can work well in the application of broadcast monitoring. (4 pages)

An audio fingerprinting system based on spectral energy structure

Audio fingerprint is also named audio hashing, which is a content-based identification technology that can be used to label an unknown audio recording Here we introduce an audio fingerprinting algorithm based on harmonic enhancement and SSC (spectral subband centroid) First, the audio signal is processed by harmonic enhancement to extract predominant pitches that are most audible for human auditory system Then SSC features of audio are computed and audio fingerprints are produced accordingly Preliminary experimental results suggest that the proposed audio fingerprinting algorithm can achieve good recognition accuracy and is possible to be used in applications such as broadcast monitoring and audio excerpt identification by mobile devices

Audio fingerprinting based on harmonic enhancement and spectral subband centroid

In the time of digital information, audio data has become an important part in many modern computer applications. Automatic classification based on audio content has been considered as an important way to cope with the problem of audio structuration. In this paper, we present an improved algorithm based on FRNNC (Fuzzy-rough nearest neighbour clustering), which derive from FRNN algorithm and have combined clustering algorithm. In our work, we extract audio features from the MDCT domain and form feature vectors by introduce the concept of feature granularity, and then apply the FRNNC algorithm in audio classification. The experimental results show that our classification method not only greatly reduces the processing time of classification, but also improves the classification accuracy.

Jijun Deng

Papers

Audio fingerprinting based on local energy centroid

Audio fingerprinting based on spectral energy structure and NMF

An audio fingerprinting system based on spectral energy structure

Audio fingerprinting based on harmonic enhancement and spectral subband centroid

Audio classification based on fuzzy-rough nearest neighbour clustering