Signal modeling techniques in speech recognition

Abstract

We have seen three important trends develop in the last five years in speech recognition. First, heterogeneous parameter sets that mix absolute spectral information with dynamic, or timederivative, spectral information, have become common. Second, similariry transform techniques, often used to normalize and decorrelate parameters in some computationally inexpensive way, have become popular. Third, the signal parameter estimation problem has merged with the speech recognition process so that more sophisticated statistical models of the signal's spectrum can be estimated in a closed-loop manner. In this paper, we review the signal processing components of these algorithms. These algorithms are presented as part of a unified view of the signal parameterization problem in which there are three major tasks: measurement, transformation, and statistical modeling. This paper is by no means a comprehensive survey of all possible techniques of signal modeling in speech recognition. There are far too many algorithms in use today to make an exhaustive survey feasible (and cohesive). Instead, this paper is meant to serve as a tutorial on signal processing in state-of-the-art speech recognition systems and to review those techniques most commonly used. In keeping with this goal, a complete mathematical description of each algorithm has been included in the paper.

A Review of Audio Fingerprinting

“…Some systems only use the derivative of the features, not the absolute features [7,19]. Using the derivative of the signal measurements tends to amplify noise [10] but, at the same time, filters the distortions produced in linear time invariant, or slowly varying channels (like an equalization). Cepstrum Mean Normalization (CMN) is used to reduce linear slowly varying channel distortions in [24].…”

“…In analogy to the cryptographic hash value, content-based digital signatures can be seen as evolved versions of hash values that are robust to content-preserving transformations [4,5]. Also from a pattern matching point of view, the idea of extracting the essence of a class of objects retaining its main characteristics is at the heart of any classification system [6][7][8][9][10].…”

Robust Speaker Identification System Based on Wavelet Transform and Gaussian Mixture Model

“…In the literature, some models have been proposed for representing speech features [1,[3][4][5][6][17][18][19]. In this paper, four well-known models, Linear Predict Coding Cepstrum (LPCC) [1], Fourier Transform Cepstral Coefficients Test Utterance (Sec)…”

“…(FTCC) [17], Generalized Mel Frequency Cepstral Coefficients (GMFCC) [3] and Wavelet Packet Transform (WPT) [18], are compared. These different modeling techniques are worthy of comparison because they represent different ways of modeling the acoustic feature distribution.…”

Development of speech recognition system in native language for hearing impaired

“…The types of signal models are deterministic and statistical models. In statistical model (Picone 1993) one tries to characterize the statistical properties of the signal. In HMM for each state, there is an output probability distribution of an acoustic vector, and each iteration is associated with a state-transition probability.…”