This tutorial provides an overview of the basic theory of hidden Markov models (HMMs) as originated by L.E. Baum and T. Petrie (1966) and gives practical details on methods of implementation of the theory along with a description of selected applications of the theory to distinct problems in speech recognition. Results from a number of original sources are combined to provide a single source of acquiring the background required to pursue further this area of research. The author first reviews the theory of discrete Markov chains and shows how the concept of hidden states, where the observation is a probabilistic function of the state, can be used effectively. The theory is illustrated with two simple examples, namely coin-tossing, and the classic balls-in-urns system. Three fundamental problems of HMMs are noted and several practical techniques for solving these problems are given. The various types of HMMs that have been studied, including ergodic as well as left-right models, are described. >

/pdf/a-tutorial-on-hidden-markov-models-and-selected-applications-1vlzbxbk68.pdf

A tutorial on hidden Markov models and selected applications in speech recognition

The use of hidden Markov models for speech recognition has become predominant in the last several years, as evidenced by the number of published papers and talks at major speech conferences. The reasons this method has become so popular are the inherent statistical (mathematically precise) framework; the ease and availability of training algorithms for cstimating the parameters of the models from finite training sets of speech data; the flexibility of the resulting recognition system in which one can easily change the size, type, or architecture of the models to suit particular words, sounds, and so forth; and the ease of implementation of the overall recognition system. In this expository article, we address the role of statistical methods in this powerful technology as applied to speech recognition and discuss a range of theoretical and practical issues that are as yet unsolved in terms of their importance and their effect on performance for different system implementations.

/pdf/hidden-markov-models-for-speech-recognition-1vqk5sdtvy.pdf

Hidden Markov models for speech recognition

A tutorial on signal processing in state-of-the-art speech recognition systems is presented, reviewing those techniques most commonly used. The four basic operations of signal modeling, i.e. spectral shaping, spectral analysis, parametric transformation, and statistical modeling, are discussed. Three important trends that have developed in the last five years in speech recognition are examined. First, heterogeneous parameter sets that mix absolute spectral information with dynamic, or time-derivative, spectral information, have become common. Second, similarity 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. The signal processing components of these algorithms are reviewed. >

/pdf/signal-modeling-techniques-in-speech-recognition-2s2amcdjv6.pdf

Signal modeling techniques in speech recognition

Speech recognition in noisy environments: a survey

The multimodal nature of speech is often ignored in human-computer interaction, but lip deformations and other body motion, such as those of the head, convey additional information. We integrate speech cues from many sources and this improves intelligibility, especially when the acoustic signal is degraded. The paper shows how this additional, often complementary, visual speech information can be used for speech recognition. Three methods for parameterizing lip image sequences for recognition using hidden Markov models are compared. Two of these are top-down approaches that fit a model of the inner and outer lip contours and derive lipreading features from a principal component analysis of shape or shape and appearance, respectively. The third, bottom-up, method uses a nonlinear scale-space analysis to form features directly from the pixel intensity. All methods are compared on a multitalker visual speech recognition task of isolated letters.

/pdf/extraction-of-visual-features-for-lipreading-27kly9h9bf.pdf

Extraction of visual features for lipreading

The vast majority of commercially available isolated word recognizers use a filter bank analysis as the front end processing for recognition. It is not well understood how the parameters of different filter banks (e.g., number of filters, types of filters, filter spacing, etc.) affect recognizer performance. In this paper we present results of performance evaluation of several types of filter bank analyzers in a speaker trained isolated word recognition test using dialed-up telephone line recordings. We have studied both DFT (discrete Fourier transform) and direct form implementations of the filter banks. We have also considered uniform and nonuniform filter spacings. The results indicate that the best performance (highest word accuracy) is obtained by both a 15-channel uniform filter bank and a 13-channel nonuniform filter bank (with channels spacing along a critical band scale). The performance of a 7-channel critical band filter bank is almost as good as that of the two best filter banks. In comparison to a conventional linear predictive coding (LPC) word recognizer, the performance of the best filter bank recognizers was, on average, several percent worse than that of an eighth-order LPC-based recognizer. A discussion as to why some filter banks performed better than others, and why the LPC-based system did the best, is given in this paper.

On the effects of varying filter bank parameters on isolated word recognition

Accurate location of the endpoints of spoken words and phrases is important for reliable and robust speech recognition. The endpoint detection problem is fairly straightforward for high-level speech signals in low-level stationary noise environments (e.g., signal-to-noise ratios greater than 30-dB rms). However, this problem becomes considerably more difficult when either the speech signals are too low in level (relative to the background noise), or when the background noise becomes highly nonstationary. Such conditions are often encountered in the switched telephone network when the limitation on using local dialed-up lines is removed. In such cases the background noise is often highly variable in both level and spectral content because of transmission line characteristics, transients and tones from the line and/or from signal generators, etc. Conventional speech endpoint detectors have been shown to perform very poorly (on the order of 50-percent word detection) under these conditions. In this paper we present an improved word-detection algorithm, which can incorporate both vocabulary (syntactic) and task (semantic) information, leading to word-detection accuracies close to 100 percent for isolated digit detection over a wide range of telephone transmission conditions.

An improved word-detection algorithm for telephone-quality speech incorporating both syntactic and semantic constraints

To implement an isolated word recognizer based on filter bank techniques, decisions must be made as to how to condition the speech signal prior to the filter bank analysis (preprocessing), how to condition the feature vector at the output of the filter bank analysis (postprocessing), and how to perform the time alignment in the pattern comparison between an unknown test pattern and previously stored reference patterns (registration and distance computation). In the past most designers of such word recognition systems made arbitrary choices about how the various signal processing operations were to be carried out. This paper presents results of a systematic study of the effects of selected signal processing techniques on the performance of a filter bank isolated word recognizer using telephone-quality speech. In particular, the filter bank analyzer was a 13-channel, critical-band-spaced filter bank with excellent time resolution (impulse response durations of from 3 to 30 ms) and poor frequency selectivity (highly overlapping filters with ratios of center frequency to 3-dB bandwidth of about 8 for each band). Among the signal processing techniques studied were: preemphasis of the speech signal; time and frequency smoothing of the filter bank outputs; thresholding, quantization, and normalization of the feature vector; principal components analysis of the feature vector; local and global distance computations for use in the time alignment procedure; and noise analysis in both training and testing. Each of the signal processing techniques was studied individually; hence no tests were run in which several of the techniques were used together. Results showed that some fairly simple signal processing operations provided the best overall performance in the noise-free case; in noisy conditions performance degraded significantly for signal-to-noise ratios less than about 24 dB.

The effects of selected signal processing techniques on the performance of a filter-bank-based isolated word recognizer

Ieee acoustics, speech, and signal processing society

The vast majority of commercially available isolated word recognizers use a filter bank analysis as the front end processing for recognition. To the designers of such recognizers there exists a myriad of design choices. It is not well understood how these designs choices (e.g. the number of filters, filter spacing, and post-processing of the filter bank outputs) affect recognizer performance. In this paper we present results of a performance evaluation of filter bank recognizers in a speaker-trained, isolated word recognition test using dialed-up telephone line recordings. First we studied the effects of various filter bank parameters on system performance. We designed a total of 13 filter banks including 8 uniform and 5 non-uniform filter banks. Within these 13 filter banks we also considered both slightly and highly overlapping filters. The results indicate that the best performance (highest word accuracy) on the 39 word alpha digits vocabulary, with 4 talkers, is obtained by both a 15 channel uniform filter bank and a 13 channel highly overlapping non-uniform critical band filter bank. Next we studied the effects of selected preprocessing and post-processing techniques on system performance. For this we used the non-uniform 13 channel filter bank. The results indicate that almost none of the processing techniques improved system performance; however, some techniques can potentially reduce hardware cost (computation and storage) without adversely affecting system performance. We also compared the results of the best filter banks recognizers with a conventional LPC based word recognizer on the same vocabulary. The performance of the best filter bank was approximately 4% worse than that of an 8th order LPC-based recognizer. We also studied the effect of additive wideband Gaussian noise on system performance of both the filter bank and the LPC recognizers. Gaussian white noise was added to the speech recordings at signal-to-noises ratios of from 0 to 30 dB. Recognition tests were then performed which indicated that the LPC system performance degraded faster than that of the filter banks; however, the point at which both systems have identical performance is at a signal-to-noise ratio of 6 dB.

T. Martin

Papers

On the effects of varying filter bank parameters on isolated word recognition

An improved word-detection algorithm for telephone-quality speech incorporating both syntactic and semantic constraints

The effects of selected signal processing techniques on the performance of a filter-bank-based isolated word recognizer

Ieee acoustics, speech, and signal processing society

On the use of filter bank features for isolated word recognition