Karthikeyan Umapathy

Audio Signal Feature Extraction and Classification Using Local Discriminant Bases

Audio feature extraction plays an important role in analyzing and characterizing audio content. Auditory scene analysis, content-based retrieval, indexing, and fingerprinting of audio are few of the applications that require efficient feature extraction. The key to extract strong features that characterize the complex nature of audio signals is to identify their discriminatory subspaces. In this paper, we propose an audio feature extraction and a multigroup classification scheme that focuses on identifying discriminatory time-frequency subspaces using the local discriminant bases (LDB) technique. Two dissimilarity measures were used in the process of selecting the LDB nodes and extracting features from them. The extracted features were then fed to a linear discriminant analysis-based classifier for a three-level hierarchical classification of audio signals into ten classes. In the first level, the audio signals were grouped into artificial and natural sounds. Each of the first level groups were subdivided to form the second level groups viz. instrumental, automobile, human, and nonhuman sounds. The third level was formed by subdividing the four groups of the second level into the final ten groups (drums, flute, piano, aircraft, helicopter, male, female, animals, birds and insects). A database of 213 audio signals were used in this study and an average classification accuracy of 83% for the first level (113 artificial and 100 natural sounds), 92% for the second level (73 instrumental and 40 automobile sounds; 40 human and 60 nonhuman sounds), and 89% for the third level (27 drums, 15 flute, and 31 piano sounds; 23 aircraft and 17 helicopter sounds; 20 male and 20 female speech; 20 animals, 20 birds and 20 insects sounds) were achieved. In addition to the above, a separate classification was also performed combining the LDB features with the mel-frequency cepstral coefficients. The average classification accuracies achieved using the combined features were 91% for the first level, 99% for the second level, and 95% for the third level

Multigroup classification of audio signals using time-frequency parameters

The ongoing advancements in the multimedia technologies drive the need for efficient classification of the audio signals to make the content-based retrieval process more accurate and much easier from huge databases. The challenge of this task lies in an accurate extraction of signal characteristics so as to derive a strong discriminatory feature suitable for classification. In this paper, a time-frequency (TF) approach for audio classification is proposed. Audio signals are nonstationary in nature and TF approach is the best way to analyze them. The audio signals were decomposed using an adaptive TF decomposition algorithm, and the signal decomposition parameter based on octave (scaling) was used to generate a set of 42 features over three frequency bands within the auditory range. These features were analyzed using linear discriminant functions and classified into six music groups (rock, classical, country, jazz, folk and pop). Overall classification accuracies as high as 97.6 % was achieved by linear discriminant analysis of 170 audio signals.

Modified local discriminant bases algorithm and its application in analysis of human knee joint vibration signals

Knee joint disorders are common in the elderly population, athletes, and outdoor sports enthusiasts. These disorders are often painful and incapacitating. Vibration signals [vibroarthrographic (VAG)] are emitted at the knee joint during the swinging movement of the knee. These VAG signals contain information that can be used to characterize certain pathological aspects of the knee joint. In this paper, we present a noninvasive method for screening knee joint disorders using the VAG signals. The proposed approach uses wavelet packet decompositions and a modified local discriminant bases algorithm to analyze the VAG signals and to identify the highly discriminatory basis functions. We demonstrate the effectiveness of using a combination of multiple dissimilarity measures to arrive at the optimal set of discriminatory basis functions, thereby maximizing the classification accuracy. A database of 89 VAG signals containing 51 normal and 38 abnormal samples were used in this study. The features extracted from the coefficients of the selected basis functions were analyzed and classified using a linear-discriminant-analysis-based classifier. A classification accuracy as high as 80% was achieved using this true nonstationary signal analysis approach.

Audio signal feature extraction and classification using local discriminant bases

Automatic classification of audio signals is an interesting and a challenging task. With the rapid growth of multimedia content over Internet, intelligent content-based audio and video retrieval techniques are required to perform efficient search over vast databases. Classification schemes form the basis of such content-based retrieval systems. In this paper we propose an audio classification scheme using local discriminant bases (LDB) algorithm. The audio signals were decomposed using wavelet packets and the high discriminatory nodes were selected using the LDB algorithm. Two different dissimilarity measures were used to select the LDB nodes and to extract features from them. The features were fed to a linear discriminant analysis based classifier for a six group (Rock, Classical, Country, Folk, Jazz and Pop) and a four group (Rock, Classical, Country and Folk) classifications. Overall classification accuracies as high as 77% and 88% were achieved for the six and four group classifications respectively using a database of 170 audio signals.

Feature analysis of pathological speech signals using local discriminant bases technique

Speech is an integral part of the human communication system. Various pathological conditions affect the vocal functions, inducing speech disorders. Acoustic parameters of speech are commonly used for the assessment of speech disorders and for monitoring the progress of the patient over the course of therapy. In the last two decades, signal-processing techniques have been successfully applied in screening speech disorders. In the paper, a novel approach is proposed to classify pathological speech signals using a local discriminant bases (LDB) algorithm and wavelet packet decompositions. The focus of the paper was to demonstrate the significance of identifying the signal subspaces that contribute to the discriminatory characteristics of normal and pathological speech signals in a computationally efficient way. Features were extracted from target subspaces for classification, and time-frequency decomposition was used to eliminate the need for segmentation of the speech signals. The technique was tested with a database of 212 speech signals (51 normal and 161 pathological) using the Daubechies wavelet (db4). Classification accuracies up to 96% were achieved for a two-group classification as normal and pathological speech signals, and 74% was achieved for a four-group classification as male normal, female normal, male pathological and female pathological signals.

Sub-dictionary selection using local discriminant bases algorithm for signal classification

In signal decompositions using overcomplete, redundant time-frequency (TF) dictionaries, often it is challenging to restrict the dictionary to a sub-dictionary tailored for specific applications. In the proposed technique we used a similar approach as local discriminant bases algorithm (LDB) to select optimal TF subdictionaries for signal classification applications. A novel time-width versus frequency band mapping was generated for each of the signal class. These mappings of different classes were compared using a discriminant measure to arrive at a sub-dictionary. This sub-dictionary was then used for decomposing the testing set signals, followed by feature extraction and classification. Two highly nonstationary biomedical databases (1) vibroarthrographic signals (89 signals, 51 normal and 38 abnormal) (2) pathological speech database (100 signals, 50 normal and 50 pathological) were tested. Classification accuracies as high as 74.2% and 92% were achieved respectively. Due to the sub-dictionary approach, approximately a 40% reduction in signal decomposition time was observed for the tested databases.

A signal classification approach using time-width vs frequency band sub-energy distributions

Time-frequency (TF) signal decompositions provide us with ample information and extreme flexibility for signal analysis. By applying suitable processing on the TF decomposition parameters, even subtle signal characteristics can be revealed. In many real world applications, identification of these subtle differences make a significant impact in signal analysis. Particularly in classification applications using TF approaches, there may be situations where a localized high discriminative signal structure is diluted due to the presence of other overlapping signal structures. To address this problem we propose a novel approach to construct multiple time-width vs frequency band mappings based on the energy decomposition pattern of the signal. These mapping are then analyzed to locate the highly discriminative features for classification. Initial results with two real world biomedical signal databases: (1) vibroarthrographic (VAG) signals; and (2) pathological speech signals, indicate high potential for the proposed technique.

Modified local discriminant bases and its applications in signal classification [biomedical signal examples]

One of the major challenges in classification problems, based on the signal decomposition approach, is to identify the right basis function and its derivatives that can provide optimal features to distinguish the classes. With the vast amount of available libraries of orthonormal bases, it is hard to select an optimal set of basis functions for a specific dataset. To address this problem, pruning algorithms based on certain selection criteria, are needed. The local discriminant bases (LDB) algorithm is one such algorithm, which efficiently selects a set of significant basis functions from the library of orthonormal bases based on a certain defined dissimilarity measure. The selection of this dissimilarity measure is critical as they indirectly contribute to the performance accuracy of the LDB algorithm. In this paper, we study the impact of the dissimilarity measures on the performance of the LDB algorithm with two classification examples. Two biomedical signal databases used are: 1) vibroarthographic signals (VAG) - 89 signals with 51 normal and 38 abnormal; and 2) pathological speech signals - 100 signals with 50 normal and 50 pathological. Classification accuracies of 76.4% with the VAG database and 96% with the pathological speech database were obtained. This modified method of signal analysis using LDB has shown its powerfulness in analyzing non-stationary signals.

A general perceptual tool for evaluation of audio codecs

Subjective evaluation forms an important part of any research work, where the feedback and perception of general public or a trained set of specialists are mandatory. Many audio and video coding techniques have emerged to tackle the bandwidth problems imposed by the Internet with data compression schemes either with lossless or perceptually lossless quality. In order to evaluate the performance of these techniques a Mean Opinion Score (MOS) test has to be performed with wide variety of subjects. In this paper we present a MOS tool developed to evaluate the audio codecs both in controlled and uncontrolled listening environments. The technique is based on the international telecommunication union - radiocommunication sector (ITU-R) standard. This novel approach of performing distributed listening tests in uncontrolled environments will help researchers to collect substantial feedback and perform statistical analysis of an audio codecs performance in an efficient manner particularly for Internet driven applications. Results of perceptual evaluation of 8 sample audio files of different varieties with an adaptive time-frequency transform (ATFT) audio codec indicates the ease, independency, and the effectiveness of performing MOS studies with the proposed technique.