Filter Bank

An investigation into the feature extraction and selection of infant cry with asphyxia is presented in this paper. The feature of the cry signal was extracted using mel frequency cepstrum coefficient (MFCC) analysis and the significant coefficients were selected using orthogonal least square (OLS) algorithm. The effect of varying the number of MFCC filter banks on the feature selection was examined. It was found that the best set of coefficients could be achieved when 40 filter banks were used.

Filter banks with fixed time-frequency resolution, such as the Short-Time Fourier Transform (STFT), are a common tool for many audio analysis and processing applications allowing effective implementation via the Fast Fourier Transform (FFT). The fixed time-frequency resolution of the STFT can lead to the undesirable smearing of events in both time and frequency. In this paper, we suggest adaptively varying STFT time-frequency resolution in order to reduce filter bank-specific artifacts while retaining adequate frequency resolution. Several strategies for systematic adaptation of time-frequency resolution are proposed. The introduced approach is demonstrated as applied to spectrogram displays, noise reduction, and spectral effects processing.

Abstract-A novel,approach,is proposed,for signal matched,DFT filter bank. To obtain signal matched,DFT filter bank we first pass the given signal through DFT filter bank so as,to decompose,the signal into M distinct uniform bands, and then each channel is passed through a filter which is a whitening,filter for that band. Since the decomposition,and whitening de-correlates within and also across the band, information

In this paper, the AM–FM modulation model is applied to speech analysis, synthesis and coding. The AM–FM model represents the speech signal as the sum of formant resonance signals each of which contains amplitude and frequency modulation. Multiband filtering and demodulation using the energy separation algorithm are the basic tools used for speech analysis. First, multiband demodulation analysis (MDA) is

The aim of this thesis is to define a perceptual scale for the ‘Time-Frequency’ analysis of music signals. The equal tempered ‘Bach’ scale is a suitable scale, since it covers most of the genres of music and the error is equally distributed for each semi-tone. However, it may be necessary to allow a tolerance of around 50 cents or half the interval of the Bach scale, so that the interval can accommodate other common intonation schemes. The thesis covers the formulation of the Bach scale filter-bank as a time-varying model. It makes a comparative study with other commonly used perceptual scales. Two applications for the Bach scale filter-bank are also proposed, namely automated segmentation of speech signals and transcription of singing voice for query-by-humming applications.

Even though this filter-bank is suggested with a motivation from music, it could also be applied to speech. A method for automatically segmenting continuous speech into phonetic units is proposed. The results, obtained from the proposed method, show around 82% accuracy for the English and 85% accuracy for the Hindi databases. This is an improvement of around 2 - 3% when the performance is compared with other popular methods in the literature. Interestingly, the Bach scale filters perform better than the filters designed for other common perceptual scales, such as Mel and Bark scales.

‘Musical transcription’ refers to the process of converting a musical rendering or performance into a set of symbols or notations. A query in a ‘query-by-humming system’ can be made in several ways, some of which are singing with words, or with arbitrary syllables, or whistling. Two algorithms are suggested to annotate a query. The algorithms are designed to be fairly robust for these various forms of queries. The first algorithm is a frequency selection based method. It works on the basis of selecting the most likely frequency components at any given time instant. The second algorithm works on the basis of finding time-connected contours of high energy in the ‘Time-Frequency’ plane of the input signal. The time domain algorithm works better in terms of instantaneous pitch estimates. It results in an error of around 10 - 15%, while the frequency domain method results in an error of around 12 - 20%.

A song rendered by two different people will have quite a few different properties. Their absolute pitches, rates of rendering, timbres based on voice quality and inaccuracies, may be different. The thesis discusses a method to quantify the distance between two different renderings of musical pieces. The distance function has been evaluated by attempting a search for a particular song from a database of a size of 315, made up of songs sung by both male and female singers and whistled queries. Around 90 % of the time, the correct song is found among the top five best choices picked.

Thus, the Bach scale has been proposed as a suitable scale for representing the perception of music. It has been explored in two applications, namely automated segmentation of speech and transcription of singing voices. Using the transcription obtained, a measure of the distance between renderings of musical pieces has also been suggested.

This paper concentrates on an efficient prototype filter de- sign in the context of filter bank based multicarrier (FBMC) transmission. An advantage of the chosen method, frequency sampling technique, is that near perfect reconstruction pro- totype filters can be expressed using a closed-form represen- tation with only a few adjustable parameters. The perfor- mance of various designs are analyzed using

Development of Text to Speech (TTS) systems in Indian languages has been difficult owing to the non availability of sizable segmented speech database of good quality. Further, no prosody models exist of most of the Indian languages. We have created comprehensive databases in Hindi, Kannada and Tamil by recording utterances covering various acoustic and phonetic contexts at the paragraph level and monologues to also facilitate super-sentential prosodic modeling of those languages.
The challenge of automating segmentation of such long sentences has been addressed in this thesis. We propose two explicit methods of segmentation and also try to accomplish language independence.
To overcome the major disadvantage of boundary error accumulation due to forcing boundary alignments on the entire (continuous) speech waveform, we propose to segment speech between two successive stop consonants. The stop consonants are detected based on the Euclidean distance between the MFCCs of any frame and MFCC of the first frame. These thus detected stop consonants acts as anchor points for the second stage of the segmentation algorithms proposed.
Both the algorithms make use of the distance function obtained from the output of the Bach scale filter bank (formulated by ‘music’ related criteria). The first explicit method proposed is a 3-stage algorithm, which uses Quatieri's sinusoidal model for speech in conjunction with a distance function obtained from Bach scale filter bank, to force align the boundaries of phonemes between 2 stop consonants. The second method proposed makes use of the distance function obtained from the output of the Bach scale filter bank and the statistical knowledge of the lengths of the phonemes for the explicit alignment. The results for both the methods in terms of frame error rate are encouraging and have been tested on Hindi, Kannada, Tamil and TIMIT (English) databases.

We investigate the application of Support Vector Machines (SVMs) in computer vision. SVM is a learning technique developed by V. Vapnik and his team (AT&T Bell Labs., 1985) that can be seen as a new method for training polynomial, neural network, or Radial Basis Functions classifiers. The decision surfaces are found by solving a linearly constrained quadratic programming problem. This optimization problem is challenging because the quadratic form is completely dense and the memory requirements grow with the square of the number of data points. We present a decomposition algorithm that guarantees global optimality, and can be used to train SVM's over very large data sets. The main idea behind the decomposition is the iterative solution of sub-problems and the evaluation of optimality conditions which are used both to generate improved iterative values, and also establish the stopping criteria for the algorithm. We present experimental results of our implementation of SVM, and demonstrate the feasibility of our approach on a face detection problem that involves a data set of 50,000 data points