In this paper, we consider the problem of quantifying the amount of information contained in a set of features, to discriminate between various classes. We explore these ideas in the context of a speech recognition system, where an important classification sub-problem is to predict the phonetic class, given an observed acoustic feature vector. The connection between information content and speech recognition system performance is first explored in the context of various feature extraction schemes used in speech recognition applications. Subsequently, the idea of optimizing the information content to improve recognition accuracy is generalized to a linear projection of the underlying features. We show that several prior methods to compute linear transformations (such as linear/heteroscedastic discriminant analysis) can be interpreted in this general framework of maximizing the information content. Subsequently, we extend this reasoning and propose a new objective function to maximize a penalized mutual information (pMI) measure. This objective function is seen to be very well correlated with the word error rate of the final system. Finally experimental results are provided that show that the proposed pMI projection consistently outperforms other methods for a variety of cases, leading to relative improvements in the word error rate of 5%-16% over earlier methods.

In this paper, we describe an adaptation method for speech recognition systems that is based on a nonlinear transformation of the feature space. In contrast to most existing adaptation methods which assume some form of affine transformation of either the feature vectors or the acoustic models that model the feature vectors, our proposed method composes a general nonlinear transformation from two transformations, one of these being an affine transformation that combines the dimensions of the original feature space, and the other being a nonlinear transformation that is applied independently to each dimension of the previously transformed feature space leading to a general multidimensional nonlinear transformation of the original feature space. This method also differs from other affine techniques in the way the parameters of the transform are shared. In most previous methods, the parameters of the transformation are shared on the basis of the phonetic class, in our method, the parameters of the nonlinear transformation are shared not on the basis of the phonetic class, but rather on the location in the feature space. Experimental results show that the method outperforms affine methods providing up to a 25% relative improvement in the word error rate in an in-car speech recognition task.

We report on the system IBM fielded in the second SPeech In Noisy Environments (SPINE-2) evaluation, conducted by the Naval Research Laboratory in October 2001. The key components of the system include an HMM-based automatic segmentation module using a novel set of LDA-transformed voicing and energy features, a multiple-pass decoding strategy that uses several speaker-and environment-normalization operations to deal with the highly variable acoustics of the evaluation, the combination of hypotheses from decoders operating on three distinct acoustic feature sets, and a class-based language model that uses both the SPINE-1 and SPINE-2 training data to estimate reliable probabilities for the new SPINE-2 vocabulary.

We report on the performance of automatic speech recognition (ASR) systems on voicemail transcription. Voicemail is spontaneous telephone speech recorded over a variety of channels; consequently, it is representative of many challenging problems in speech recognition. In the course of working on this task, several algorithms were developed that focus on different components of an ASR system, including lexicon design, feature extraction, hypothesis search, and adaptation. We report the improvements provided by these techniques, as well as other standard techniques, on a voicemail test set. Although the techniques are benchmarked on voicemail test data, their scope is not restricted to this domain as they address fundamental aspects of the speech recognition process.

In this paper we address the problem of extracting key pieces of information from voicemail messages, such as the identity and phone number of the caller. This task differs from the named entity task in that the information we are interested in is a subset of the named entities in the message, and consequently, the need to pick the correct subset makes the problem more difficult. Also, the caller's identity may include information that is not typically associated with a named entity. In this work, we present three information extraction methods, one based on hand-crafted rules, one based on maximum entropy tagging, and one based on probabilistic transducer induction. We evaluate their performance on both manually transcribed messages and on the output of a speech recognition system.

In the standard MAP approach to speech recognition, the goal is to find the word sequence with the highest posterior probability given the acoustic observation. A number of alternate approaches have been proposed for directly optimizing the word error rate, the most commonly used evaluation criterion. One of them, the consensus decoding approach, converts a word lattice into a confusion network which specifies the word-level confusions at different time intervals, and outputs the word with the highest posterior probability from each word confusion set. The paper presents a method for discriminating between the correct and alternate hypotheses in a confusion set using additional knowledge sources extracted from the confusion networks. We use transformation-based learning for inducing a set of rules to guide a better decision between the top two candidates with the highest posterior probabilities in each confusion set. The choice of this learning method is motivated by the perspicuous representation of the rules induced, which can provide insight into the cause of the errors of a speech recognizer. In experiments on the Switchboard corpus, we show significant improvements over the consensus decoding approach.

We extend the well-known technique of constrained maximum likelihood linear regression (MLLR) to compute a projection (instead of a full rank transformation) on the feature vectors of the adaptation data. We model the projected features with phone-dependent Gaussian distributions and also model the complement of the projected space with a single class-independent, speaker-specific Gaussian distribution. Subsequently, we compute the projection and its complement using maximum likelihood techniques. The resulting ML transformation is shown to be equivalent to performing a speaker-dependent heteroscedastic discriminant (or HDA) projection. Our method is in contrast to traditional approaches which use a single speaker-independent projection, and execute speaker adaptation in the resulting subspace. Experimental results on Switchboard show a 3% relative improvement in the word error rate over constrained MLLR in the projected subspace only.

Abstract In this paper, we describe an adaptation method for speechrecognition systems that is based on a piecewise-linear approx-imation to a non-linear transformation of the feature space.The method extends a previously proposed non-linear trans-formation (NLT) technique by making the transformation func-tion more sophisticated (piecewise-linear instead of piecewise-constant), and by computing the transformation to maximizethe likelihood of the adaptation data given its transcription (in-stead of just matching the global statistics of the test and train-ing data). This method also differs from other linear techniques(such as MLLR, linear feature space transforms, etc.) in twoways - first, the computed transformation is non-linear, second,the tying structure of the transformation depends not on the pho-netic class but rather on the location in the feature space. Exper-imental results show that the method performs well for the caseof limited adaptation data, and the performance gains appear tobe additive to those provided by MLLR - yielding upto 3.4%relative improvement over MLLR.

We present a new approach to deriving compound words from a training corpus. The motivation for making compound words is because under some assumptions, speech recognition errors occur less frequently in longer words. Furthermore, they also enable more accurate modeling of pronunciation variability at the boundary between adjacent words in a continuously spoken utterance. We introduce a measure based on the product between the direct and the reverse bigram probability of a pair of words for finding candidate pairs in order to create compound words. Our experimental results show that by augmenting both the acoustic vocabulary and the language model with these new tokens, the word recognition accuracy can be improved by absolute 2.8% (7% relative) on a voice mail continuous speech recognition task. We also compare the proposed measure for selecting compound words with other measures that have been described in the literature.

Linear discriminant analysis (LDA) is known to be inappropriate for the case of classes with unequal sample covariances. There has been an interest in generalizing LDA to heteroscedastic discriminant analysis (HDA) by removing the equal within-class covariance constraint. This paper presents a new approach to HDA by defining an objective function which maximizes the class discrimination in the projected subspace while ignoring the rejected dimensions. Moreover, we investigate the link between discrimination and the likelihood of the projected samples and show that HDA can be viewed as a constrained ML projection for a full covariance Gaussian model, the constraint being given by the maximization of the projected between-class scatter volume. It is shown that, under diagonal covariance Gaussian modeling constraints, applying a diagonalizing linear transformation (MLLT) to the HDA space results in increased classification accuracy even though HDA alone actually degrades the recognition performance. Experiments performed on the Switchboard and Voicemail databases show a 10%-13% relative improvement in the word error rate over standard cepstral processing.

In this paper, we apply boosting to the problem of frame-level phone classification, and use the resulting system to perform voicemail transcription. We develop parallel, hierarchical, and restricted versions of the classic AdaBoost algorithm, which enable the technique to be used in large-scale speech recognition tasks with hundreds of thousands of Gaussians and tens of millions of training frames. We report small but consistent improvements in both frame recognition accuracy and word error rate.

This paper describes a computationally inexpensive, nonlinear feature transformation technique for rapid adaptation of a speech recognition system to new acoustic conditions. One of the advantages of the method is that it does not require any initial decoding of the adaptation data for computing the nonlinear transform. This technique performs as well as the more expensive unsupervised MLLR technique. Furthermore, it signi cantly adds to the improvement when combined with unsupervised MLLR. 1.

In this paper we report recent improvements in word error performance on a voicemail transcription task. Last year, the speaker independent word error rate (WER) on the dev test set of the Voicemail Transcription task was reported at 35.45% [1]. This year, we report a relative 20% gain over this number. The improvements were obtained using several new algorithms and an increased amount of training data. In addition to benchmarking the performance of these algorithms on the Voicemail task, we have also evaluated them on the Switchboard task, and we report these results here as well. Finally, we also present the result of crossdomain experiments to evaluate the domain-independence of the constructed systems. 1. INTRODUCTION In this paper we report recent improvements in transcribing conversational telephone speech, as typified by the Voicemail and Switchboard transcription tasks. These improvements are a result of some new algorithms and, in the case of Voicemail, also due to an increa...

We consider the problem of designing a linear transformation 2 IR pn , of rank p n, which projects the features of a classier x 2 IR n onto y = x 2 IR p such as to achieve minimum Bayes error (or probability of misclassication). Two avenues will be explored: the rst is to maximize the -average divergence between the class densities and the second is to minimize the union Bhattacharyya bound in the range of . While both approaches yield similar performance in practice, they outperform standard LDA features and show a 10% relative improvement in the word error rate over state-of-the-art cepstral features on a large vocabulary telephony speech recognition task. 1 Introduction Modern speech recognition systems use cepstral features characterizing the short-term spectrum of the speech signal for classifying frames into phonetic classes. These features are augmented with dynamic information from the adjacent frames to capture transient spectral events in the signal. What ...

The classical dynamic programming recursions for the forwards-backwards and Viterbi HMM algorithms are linear in the number of time frames being processed. Adapting the method of [8] to the context of speech recognition, this paper uses a recursive divide-andconquer algorithm to reduce the space requirement to logarithmic in the number of frames. With this procedure, it is possible to do exact computations for observation sequences of essentially arbitrary length. The procedure works by manipulating a stack of alpha vectors, and by using sparse vectors, the space savings can be combined with those of traditional pruning techniques. We apply this technique to MAP lattice construction, and present the first results in the literature for that technique. We find that it is an effective way of creating word lattices, and that doing the exact computations enabled by the log-space technique results in lower word error rates than space saving via traditional pruning. 1 Introduction Hidden Ma...

We consider the problem of designing a linear transformation () E lRPx n, of rank p ~ n, which projects the features of a classifier x E lRn onto y = ()x E lRP such as to achieve minimum Bayes error (or probabil(cid:173) ity of misclassification). Two avenues will be explored: the first is to maximize the ()-average divergence between the class densities and the second is to minimize the union Bhattacharyya bound in the range of (). While both approaches yield similar performance in practice, they out(cid:173) perform standard LDA features and show a 10% relative improvement in the word error rate over state-of-the-art cepstral features on a large vocabulary telephony speech recognition task.

We present a discriminant measure that can be used to determine the model complexity in a speech recognition system. In the speech recognition process, sub-phonetic classes are modelled as mixtures of Gaussians, and we present a new discriminant measure that uses the classification accuracy to determine in an objective fashion, the number of Gaussians required to best model the PDF of an allophone class. We compare the performance of this criterion with other criteria such as the Bayesian information criterion (BIC), and show that the BIC and the discriminative criterion lead to parsimonious models that provide the same word error rate performance as much larger baseline systems. However, this performance improvement depends on the size of the system, and there appears to be a crossover point beyond which both the BIC and the discriminative criterion are worse than a much simpler criterion. The discriminative criterion also enables this crossover point to be controlled by means of a threshold that is used in the criterion, and can lead to a better tradeoff of complexity versus word error rate.

Two discriminant measures for HMM states to improve the effectiveness on HMM training are presented. In HMM based speech recognition, the context-dependent states are usually modeled by Gaussian mixture distributions. In general, the number of Gaussian mixtures for each state is fixed or proportional to the amount of training data. From our study, some of the states are "non-aggressive" compared to others, and a higher acoustic resolution is required for them. Two methods are presented in this paper to determine those non-aggressive states. The first approach uses the recognition accuracy of the states and the second method is based on a rank distribution of states. Baseline systems, trained by a fixed number of Gaussian mixtures for each state, having 33 K and 120 K Gaussians, yield 14.57% and 13.04% word error rates, respectively. Using our approach, a 38 K Gaussian system was constructed that reduces the error rate to 13.95%. The average ranks of non-aggressive states in rank lists of testing data were also seen to dramatic improve compared to the baseline systems.

We describe our development work on a telephonebased conversational system in the domain of mutual fund transactions. This system uses several components including robust large vocabulary continuous speech recognition, natural language understanding, dialog management, and text-to-speech synthesis technologies.

Speaker adaptation techniques have emerged as very effective and practical methods to improve ASR performance on a test speaker with only limited speech data from the speaker. We explore the use of adaptation techniques on a new Voicemail database and present some theoretical extensions of the Cluster Transformation (CT) technique. Our experiments on 40 hours of voicemail data and four clusters shows that using cluster information with MLLR improves over baseline MLLR by 2.2% (relative). When the amount of adaptation data in a short message is insufficient to reliably decide its cluster, higher improvements result when we use MLLR for the very short messages and CT on longer ones. 1. INTRODUCTION Speaker adaptation techniques have emerged as very effective and practical methods to improve ASR performance on a test speaker with the use of a limited amount of speech data from the speaker. Most of these techniques generally start with speaker-independent (SI) acoustic models, and adapt ...

In this paper we report recent improvements in voicemail transcription. Last year, the speaker independent and speaker adapted word error rates (WER) on the Voicemail Transcription task were reported at 41.94% and 38.18% respectively. This year, we report a relative improvement of 18% in the speaker independent performance and 11% in the speaker adapted performance over last year. This improvement is a result of some new algorithms and an increase in the amount of training data. In the following sections, we describe the contribution of several components to improving the word error rate. 1. INTRODUCTION In this paper we report recent improvements in voicemail transcription. The voicemail transcription task was introduced last year [1] as representing a style of conversational telephone speech that is somewhat different from the Switchboard and CallHome databases. Last year, the speaker independent and speaker adapted word error rates (WER) on this task were reported at 41.94% and 38...

In this paper we apply Bayesian networks to the problem of voicemail transcription. We use a Bayesian network system to test a variety of probabilistic models that model acoustic context in addition to phonetic state and acoustic observations. We use a context variable that has the ability to model contextual phenomena that are not implied by the linguistic sequence of phones (e.g. noise level or speech rate). In rescoring experiments, we are able to get a slight gain over a more standard system with a similar number of parameters. We obtained the best performance by conditioning the mixture coefficients on context, thus implementing a state-wise tied mixture system. In an utteranceclustering system, analysis of the learned parameters indicates that the context variable is highly correlated with and . 1. INTRODUCTION The Bayesian network formalism combines a graphical way of representing probabilistic models with a set of very general algorithms for computing with these mo...

We describe the design and use of linear hyperplanes to partition the feature space of a classifier. The objective of the partitioning is to minimize the average entropy of the class distribution in the final partitions. The hyperplanes are characterized by a vector /spl nu//sub n/ and scalar h/sub n/, which are computed with the objective of maximizing the mutual information associated with the partitioning. We show that the problem of designing the /spl nu//sub n/ can be simplified to an approximately equivalent problem, the solution of which, interestingly enough, turns out to be the linear discriminant of the data. We also describe a decision-tree based technique that partitions the feature space in a hierarchical fashion where each node in the decision tree represents a linear hyperplane that further partitions the feature space into two regions. The end result of the tree-growing process is that the entire feature space is partitioned into nonoverlapping regions where each region is bounded by a number of hyperplanes. As the criterion for the design of the hyperplanes is the minimization of the average class entropy of the regions, each region is characterized by the occurrence of only a small number of classes. The partitioning information provided by the decision-tree can then be used to eliminate a number of classes from being considered, and hence simplifies the job of the classifier. We show the application of the decision-tree as a preprocessor to a classifier in the speech recognition problem. Here the classes are modeled with mixtures of tens to hundreds of thousands of Gaussians, and the use of the partitioning information reduces the computation associated with the classification process by factors larger than 20 with negligible degradation in the word error rate (note, however, that the overall decoding process comprises other steps in addition to the classification-the computational speedup mentioned above does not affect the computation of these other steps). We also compare the performance of this quantization scheme with standard Gaussian clustering schemes and show that the decision-tree based quantization provides better performance in most cases.

We present a discriminant measure that can be used to determine the model complexity in a speech recognition system. In the speech recognition process, given a test feature vector the conditional probability of the feature vector has to be obtained for several allophone (sub-phonetic units) classes using a Gaussian-mixture density model for each class. The Gaussian-mixture models are constructed from the training data belonging to the allophone classes, and the number of mixture components that are required to adequately model the PDF of each class is determined by using some simple rule of thumb-for instance the number of components has to be sufficient to model the data reasonably well but not so many as to overmodel the data. A typical example of the choice of the number is to make it proportional to the number of data samples. However, such methods may result in models that are sub-optimal as far as classification accuracy is concerned. We present a new discriminant measure that can be used to determine in an objective fashion, the number of Gaussians required to best model the PDF of an allophone class. We also present the results of experiments showing the improvement in recognition performance when the number of mixture components is chosen based on the discriminant measure as opposed to the rule of thumb. These results are presented both for the speaker-independent and speaker-adapted case.

We describe a new testbed for developing speech recognition algorithms-the ARRPA-sponsored voicemail transcription task, analogous to other tasks such as the Switchboard, CallHome and the Hub 4 tasks. The task involves the transcription of voicemail conversations. Voicemail represents a very large volume of real-world speech data, which is however not particularly well represented in existing databases. For instance, the Switchboard and CallHome databases contain telephone conversations between two humans, representing telephone-bandwidth spontaneous speech; the Hub 4 database contains radio broadcasts which represents different kinds of speech data such as spontaneous speech from a well-trained speaker, conversations between two humans possibly over the telephone, etc. The voicemail database on the other hand also represents telephone bandwidth spontaneous speech, however the difference with respect to the Switchboard and CallHome tasks is that the interaction is not between two humans, but rather between a human and a machine-consequently, the speech is expected to be a little more formal in its nature, without the problems of crosstalk, barge-in etc. This eliminates some of the variables and provides more controlled conditions enabling one to concentrate on the aspects of spontaneous speech and effects of the telephone channel. We describe the modality of collection of the speech data, and some algorithmic techniques that were devised based on this data. We also describe the initial results of the transcription performance on this task.

Phonetic baseforms are the basic recognition units in most speech recognition systems. These baseforms are usually determined by linguists once a vocabulary is chosen and not modified thereafter. However, several applications, such as name dialing, require the user be able to add new words to the vocabulary. These new words are often names, or task-specific jargon, that have user-specific pronunciations. This paper describes a novel method for generating phonetic transcriptions (baseforms) of words based on acoustic evidence alone. It does not require either the spelling or any prior acoustic representation of the new word, is vocabulary independent, and does not have any linguistic constraints (pronunciation rules). Our experiments demonstrate the high decoding accuracies obtained when baseforms deduced using this approach are incorporated into our speech recognizer. Also, the error rates on the added words were found to be comparable to or better than when the baseforms were derived by hand.

A speaker adaptation strategy is described that is based on finding a subset of speakers, from the training set, who are acoustically close to the test speaker, and using only the data from these speakers (rather than the complete training corpus) to reestimate the system parameters. Further, a linear transformation is computed for every one of the selected training speakers to better map the training speaker's data to the test speaker's acoustic space. Finally, the system parameters (Gaussian means) are reestimated specifically for the test speaker using the transformed data from the selected training speakers. Experiments showed that this scheme is capable of providing an 18% relative improvement in the error rate on a large-vocabulary task with the use of as little as three sentences of adaptation data.

We present a decision-tree based procedure to quantize the feature-space of a speech recognizer, with the motivation of reducing the computation time required for evaluating gaussians in a speech recognition system. The entire feature space is quantized into non overlapping regions where each region is bounded by a number of hyperplanes. Further, each region is characterized by the occurence of only a small number of the total alphabet of allophones (sub-phonetic speech units); by identifying the region in which a test feature vector lies, only the gaussians that model the density of allophones that exist in that region need be evaluated. The quantization of the feature space is done in a heirarchical manner using a binary decision tree. Each node of the decision tree represents a region of the feature space, and is further characterized by a hyperplane (a vector v n and a scalar threshold value hn ), that subdivides the region corresponding to the current node into two non-overlapping...

Two discriminative techniques are described (and evaluated) for estimating the parameters of the Gaussians in a large vocabulary speech-recognition system. The first technique is based on using a modification of the maximum mutual information (MMI) objective function, and appears to provide no improvement over standard ML estimation. The second technique is based on a heuristic correction of the Gaussian parameters, and is seen to give a 2-5% improvement over ML estimation.

A speaker adaptation strategy is described that is based on finding a subset of speakers, from the training set, who are acoustically close to the test speaker, and using only the data from these speakers (rather than the complete training corpus) to re-estimate the system parameters. Further, a linear transformation is computed for every one of the selected training speakers to better map the training speaker's data to the test speaker's acoustic space. Finally, the system parameters (Gaussian means) are re-estimated specifically for the test speaker using the transformed data from the selected training speakers. Experiments showed that this scheme is capable of reducing the error rate by 10-15% with the use of as little as 3 sentences of adaptation data.

In this paper we discuss various experimental results using our continuous speech recognition system on the Wall Street Journal task. Experiments with different feature extraction methods, varying amounts and type of training data, and different vocabulary sizes are reported.

Speaker adaptation typically involves customizing some existing (reference) models in order to account for the characteristics of a new speaker. This work considers the slightly different paradigm of customizing some reference data for the purpose of populating the new speaker's space, and then using the resulting (augmented) data to derive the customized models. The data augmentation technique is based on the metamorphic algorithm first proposed in Bellegarda et al. [1992], assuming that a relatively modest amount of data (100 sentences) is available from each new speaker. This contraint requires that reference speakers be selected with some care. The performance of this method is illustrated on a portion of the Wall Street Journal task.

An adaptive line enhancer is presented that is based on a resonator-based IIR filter, that uses a hyperstable update to do the adaptation. The main feature of the algorithm is that the strict positive reality (SPR) requirement for convergence associated with the hyperstable update, is satisfied by imposing some simple sufficient conditions on the input sinusoidal frequencies, which may be obtained relatively easily, numerically, even for high order cases.

This paper presents a channel-bank based phone detection algorithm, that can be used in greatly cut down the search space in the process of mapping a set of acoustic features to a phone sequence. The algorithm involves a low cost preprocessing, and relies on using local information (in time) about the acoustic features, to predict, whether or not a phone can exist over a certain time interval; hence, it may be thought of as producing a short-list of possible phones at all times, and it is only necessary to search in this short list, rather than in the entire phone alphabet, for the correct phone. The technique was applied in the acoustic fast match of the IBM speech recognition system, and resulted in a significant reduction in the execution time of the fast match.>

A frequency-sampling Fourier transform can be efficiently and robustly realized by a resonator-in-a-loop filter bank with uniformity spaced resonators. If the resonator outputs are grouped using a Lerner weighting, where alternating signs are used for each resonator output, then filter banks with greatly reduced sidelobes are obtained at the expense of wider passbands. The filters obtained by grouping the real Fourier transform outputs are linear phase and have exactly 90 degrees phase shift differences compared to those obtained by a Lerner grouping of the imaginary outputs. This allows for the realization of very accurate Hilbert transform bandpass filters.>

Some further results on modulated and extended lapped transforms are reported. Sufficient conditions to ensure perfect reconstruction for extended lapped transforms are derived. These conditions are valid for transforms of length lM and place no restriction on l; in this respect they apply to a more general case than earlier derivations that constrained l to be even. Also, a new filter structure is reported for implementing the modulated lapped transform. It has a hardware complexity of O(M) as compared to O(M log M) for other implementations.>

An infinite impulse response (IIR) adaptive filter that is efficient and can be used to analyze short data segment with high resolution and very small biases is presented. The approach taken is to run the short data segment continuously forwards and backwards in time through the adaptive filter. The proposed adaptive filter is a filter bank composed of a number of parallel digital resonators with common feedback around all of them. Because of this particular architecture, the filter is easy to reset the internal state variables of the adaptive filter each time the direction of the input data segment is changed. This allows the same data segment to be run through the adaptive filter many times, with the adaptive filter converging closer to its optimum state each time. No windowing is required, which results in higher accuracies when determining the sinusoidal components of the input data. The resonator frequencies are adapted to match the frequencies of the input data segment. When the adaptive filter has converged, the isolated input sinusoids are available at the outputs of the resonators.>

A generalized filtering structure is developed which is based on the N*N basis matrix of an arbitrary transform. The filtering structure finds application as an efficient sliding transformer of data because it has the low complexity of the associated frequency sampling structure, but has much better sensitivity properties. It can also be used as a filterbank in a multirate filtering system. The analysis filters are seen to require very few multipliers (the synthesis filters are finite impulse response with, at most, as many taps as there are channels in the filterbank) and at the same time provide approximately=40 dB attenuation of the nearest sidelobe.>

Comparisons of the performance of different types of correlator receivers and the limiter-discriminator for FSK (frequency shift keying) demodulation are presented, based on a modification of the analysis used by N. Ekanayake (1984), which allows the IF (intermediate frequency) bandlimiting to be taken into account. It is shown that for optimal values of IF bandwidth and deviation ratio, all receivers perform about the same, hence, the choice of receiver type can be made based on implementational considerations. To this end, the signal-to-interference ratio is introduced as a possible criterion for selecting the receiver type. Further, some of the correlator types considered possess the capability of being tuned, allowing them to perform better for larger values of the deviation ratio and IF bandwidth. It is shown that the conditions for optimal Pe performance of FSK are modified when sampling offset is considered.>