Automatic Subgrouping of Multitrack Audio

2017 IEEE 19th International Workshop on Multimedia Signal Processing (MMSP), 2017

Using solely the information retrieved by audio fingerprinting techniques, we propose methods to treat a possibly large dataset of user-generated audio content, that (1) enable the grouping of several audio files that contain a common audio excerpt (i.e. are relative to the same event), and (2) give information about how those files are correlated in terms of time and quality inside each event. Furthermore, we use supervised learning to detect incorrect matches that may arise from the audio fingerprinting algorithm itself, whilst ensuring our model learns with previous predictions. All the presented methods were further validated by user-generated recordings of several different concerts manually crawled from YouTube.

Pattern Recognition Letters, 2010

Audio classification typically involves feeding a fixed set of low-level features to a machine learning method, then performing feature aggregation before or after learning. Instead, we jointly learn a selection and hierarchical temporal aggregation of features, achieving significant performance gains.

In this paper we have focused on an efficient feature selection method in classification of audio files. The main objective is feature selection and extraction. We have selected a set of features for further analysis, which represents the elements in feature vector. By extraction method we can compute a numerical representation that can be used to characterize the audio using the existing toolbox. In this study Gain Ratio (GR) is used as a feature selection measure. GR is used to select splitting attribute which will separate the tuples into different classes. The pulse clarity is considered as a subjective measure and it is used to calculate the gain of features of audio files. The splitting criterion is employed in the application to identify the class or the music genre of a specific audio file from testing database. Experimental results indicate that by using GR the application can produce a satisfactory result for music genre classification. After dimensionality reduction best three features have been selected out of various features of audio file and in this technique we will get more than 90% successful classification result.

Applied Sciences, 2017

The numerical methods described in this paper can be used in the automatic creation of artificial datasets of audio mixes, as real-world mixes are both scarce and costly to produce. Such datasets can be used for a variety of applications, such as material for signal analysis, audio stimuli in psychoacoustic testing or as a population of solutions to be optimised, thus forming the basis of an automatic mixing system. Within this paper, the application of interest is testing the robustness of tempo estimation to re-mixing.

2000

Segmenting audio streams in a signicant manner and clustering sound segments objectively, is a signicantchallenge due to the nature of audio data. This paper presents some preliminary work on clustering sound segments based on frequency and harmonic characteristics. New metrics for comparing the similarity of sound segments are also devised. Keywords Multimedia Data Mining, Sound Processing, Classication, Clustering, Similarity comparison 1. INTRODUCTION Multimedia systems play an increasingly important role in our daily lives. Access to media through the Internet, and by a growing number of electronic media recording and playback devices, is inuencing and changing our lives in a profound way. Navigating through the vast amounts of multimedia data being generated is becoming an overwhelming problem. Although automated techniques for accessing electronic media are being developed, the science is still in an embryonic stage. In the area of audio-data, there has been promising ear...

Mathematical Problems in Engineering, 2015

Audio segmentation is a basis for multimedia content analysis which is the most important and widely used application nowadays. An optimized audio classification and segmentation algorithm is presented in this paper that segments a superimposed audio stream on the basis of its content into four main audio types: pure-speech, music, environment sound, and silence. An algorithm is proposed that preserves important audio content and reduces the misclassification rate without using large amount of training data, which handles noise and is suitable for use for real-time applications. Noise in an audio stream is segmented out as environment sound. A hybrid classification approach is used, bagged support vector machines (SVMs) with artificial neural networks (ANNs). Audio stream is classified, firstly, into speech and nonspeech segment by using bagged support vector machines; nonspeech segment is further classified into music and environment sound by using artificial neural networks and la...

2009 IEEE International Conference on Acoustics, Speech and Signal Processing, 2009

In the process of music content creation, a wide range of typical audio effects such as reverberation, equalization or dynamic compression are very commonly used. Despite the fact that such effects have a clear impact on the audio features, they are rarely taken into account when building an automatic audio classifier. In this paper, it is shown that the incorporation of prior knowledge of the digital media creation chain can clearly improve the robustness of the audio classifiers, which is demonstrated on a task of musical instrument recognition. The proposed system is based on a robust feature selection strategy, on a novel use of the virtual support vector machines technique and a specific equalization used to normalize the signals to be classified. The robustness of the proposed system is experimentally evidenced using a rather large and varied sound database.

Computers and Their …, 1999

The explosive increases in the amounts of audio (and multimedia) data being generated, processed, and used in several computer applications have necessitated the development of audio (and multimedia) database systems. The classi cation of audio data ...

International Journal of Artificial Intelligence & Applications

Deep Learning (DL) algorithms have shown impressive performance in diverse domains. Among them, audio has attracted many researchers over the last couple of decades due to some interesting patterns–particularly in classification of audio data. For better performance of audio classification, feature selection and combination play a key role as they have the potential to make or break the performance of any DL model. To investigate this role, we conduct an extensive evaluation of the performance of several cutting-edge DL models (i.e., Convolutional Neural Network, EfficientNet, MobileNet, Supper Vector Machine and Multi-Perceptron) with various state-of-the-art audio features (i.e., Mel Spectrogram, Mel Frequency Cepstral Coefficients, and Zero Crossing Rate) either independently or as a combination (i.e., through ensembling) on three different datasets (i.e., Free Spoken Digits Dataset, Audio Urdu Digits Dataset, and Audio Gujarati Digits Dataset). Overall, results suggest feature s...

2017

Sound effect libraries are commonly used by sound designers in a range of industries. Taxonomies exist for the classification of sounds into groups based on subjective similarity, sound source or common environmental context. However, these taxonomies are not standardised, and no taxonomy based purely on the sonic properties of audio exists. We present a method using feature selection, unsupervised learning and hierarchical clustering to develop an unsupervised taxonomy of sound effects based entirely on the sonic properties of the audio within a sound effect library. The unsupervised taxonomy is then related back to the perceived meaning of the relevant audio features.