A Maximum Likelihood Approach to Continuous Speech Recognition

Bahl, Lalit R.; Jelinek, F.; Mercer, R.L.

doi:10.1109/tpami.1983.4767370

Cited by 906 publications

(183 citation statements)

References 12 publications

Supporting

Mentioning

178

Contrasting

Unclassified

Order By: Relevance

“…We discuss them briefly in this section. From the communication theory perspective (Bahl et al, 1983), standard HMM-based ASR approach can be seen as a communication problem where noisy output of acoustic channel (i.e., sequence of acoustic likelihood vectors {v 1 , . .…”

Section: Similarities and Dissimilarities Between Kl-hmm Tied And Spmentioning

confidence: 99%

Acoustic and lexical resource constrained ASR using language-independent acoustic model and language-dependent probabilistic lexical model

Rasipuram

Magimai-Doss

2015

Speech Communication

View full text Add to dashboard Cite

One of the key challenge involved in building a statistical automatic speech recognition (ASR) system is modeling the relationship between lexical units (that are based on subword units in the pronunciation lexicon) and acoustic feature observations. To model this relationship two types of resources are needed, namely, acoustic resources (speech signals with word level transcriptions) and lexical resources (which transcribes each word in terms of subword units). Standard ASR systems typically use phonemes or phones as subword units. Not all languages have well developed acoustic resources and phonetic lexical resources. In this paper, we show that modeling of the relationship between lexical units and acoustic features can be factored into two parts through a latent variable, referred to as acoustic units, namely: (a) acoustic model that models the relationship between acoustic features and acoustic units and (b) lexical model that models the relationship between lexical units and acoustic units. Through this understanding, we elucidate that in standard hidden Markov model (HMM) based ASR system, the lexical model is deterministic (i.e., there exists an one-to-one relationship between lexical units and acoustic units), and it is the deterministic lexical model that imposes the need for well developed acoustic and lexical resources in the target language or domain when building ASR system. We then propose an approach that addresses both acoustic resource and lexical resource constraints. More specifically, in the proposed approach the acoustic model models the relationship between acoustic features and multilingual phones (acoustic units) on target language-independent data, and the lexical model models a probabilistic relationship between lexical units based on graphemes and multilingual phones on small amount of target language data. We show the potential and the efficacy of the proposed approach through experiments and comparisons with other approaches on three different ASR tasks, namely, non-native accented speech recognition, rapid development of ASR system for a new language and development of ASR system for a minority language.

show abstract

Section: Similarities and Dissimilarities Between Kl-hmm Tied And Spmentioning

confidence: 99%

Acoustic and lexical resource constrained ASR using language-independent acoustic model and language-dependent probabilistic lexical model

Rasipuram

Magimai-Doss

2015

Speech Communication

View full text Add to dashboard Cite

show abstract

“…Also, the probabilistic formulation of HMMs has provided a unified framework for scoring of hypotheses and for combining different knowledge sources. For example, the sequence of spoken words can also be modeled as the output of another statistical process (12). In this way it becomes natural to combine the HMMs for speech with the statistical models for language.…”

Section: Hidden Markov Modelsmentioning

confidence: 99%

“…A measure of how constrained a grammar is is given by its perplexity (12). Perplexity is defined as 2 raised to the power of the Shannon entropy of the grammar.…”

Section: Hidden Markov Modelsmentioning

confidence: 99%

“…In practice, sequences of words have greatly differing probabilities, and the perplexity is often much less than the vocabulary size, especially for larger vocabularies. Because grammars are estimated from a set of training data, it is often more meaningful to measure the perplexity on an independent set of data, or what is known as test-set perplexity (12). Test-set perplexity Q is obtained'by computing Q = P(W1 W2 ... WM)-1/M where w1 W2 ... WM is the sequence of words obtained by concatenating all the test sentences and P is the probability of that whole sequence.…”

Section: Hidden Markov Modelsmentioning

confidence: 99%

See 1 more Smart Citation

State of the art in continuous speech recognition.

Makhoul¹,

Schwartz²

1995

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

In the past decade, tremendous advances in the state of the art of automatic speech recognition by machine have taken place. A reduction in the word error rate by more than a factor of 5 and an increase in recognition speeds by several orders of magnitude (brought about by a combination of faster recognition search algorithms and more powerful computers), have combined to make high-accuracy, speakerindependent, continuous speech recognition for large vocabularies possible in real time, on off-the-shelf workstations, without the aid of special hardware. These advances promise to make speech recognition technology readily available to the general public. This paper focuses on the speech recognition advances made through better speech modeling techniques, chiefly through more accurate mathematical modeling of speech sounds.More and more, speech recognition technology is making its way from the laboratory to real-world applications. Recently, a qualitative change in the state of the art has emerged that promises to bring speech recognition capabilities within the reach of anyone with access to a workstation. High-accuracy, real-time, speaker-independent, continuous speech recognition for medium-sized vocabularies (a few thousand words) is now possible in software on off-the-shelf workstations. Users will be able to tailor recognition capabilities to their own applications. Such software-based, real-time solutions usher in a whole new era in the development and utility of speech recognition technology.As is often the case in technology, a paradigm shift occurs when several developments converge to make a new capability possible. In the case of continuous speech recognition, the following advances have converged to make the new technology possible: * higher-accuracy continuous speech recognition, based on better speech modeling techniques; * better recognition search strategies that reduce the time needed for high-accuracy recognition; and * increased power of audio-capable, off-the-shelf workstations. The paradigm shift is taking place in the way we view and use speech recognition. Rather than being mostly a laboratory endeavor, speech recognition is fast becoming a technology that is pervasive and will have a profound influence on the way humans communicate with machines and with each other. This paper focuses on speech modeling advances in continuous speech recognition, with an exposition of hidden Markov models (HMMs), the mathematical backbone behind these advances. While knowledge of properties of the speech signal and of speech perception have always played a role, recent improvements have relied largely on solid mathematical and probabilistic modeling methods, especially the use of HMMs for modeling speech sounds. These methods are capable of modeling time and spectral variability simultaneously, and the model parameters can be estimated automatically from given training speech data. The traditional processes of segmentation and labeling of speech sounds are now merged into a single probabilistic proce...

show abstract

“…P(A IW) is the probability that if W is uttered, data A = a1, a2, ... ak will be observed (2). In this simplified presentation the elements ai are assumed to be symbols from some finite alphabet s of size isl.…”

Section: Introductionmentioning

confidence: 99%

Training and search methods for speech recognition.

Jelinek

1995

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Speech recognition involves three processes: extraction of acoustic indices from the speech signal, estimation of the probability that the observed index string was caused by a hypothesized utterance segment, and determination of the recognized utterance via a search among hypothesized alternatives. This paper is not concerned with the first process. Estimation of the probability of an index string involves a model of index production by any given utterance segment (e.g., a word). Hidden Markov models (HMMs) are used for this purpose [Makhoul, J. & Schwartz, R. (1995) Proc.Nati. Acad. Sci. USA 92,[9956][9957][9958][9959][9960][9961][9962][9963]. Their parameters are state transition probabilities and output probability distributions associated with the transitions. The Baum algorithm that obtains the values of these parameters from speech data via their successive reestimation will be described in this paper. The recognizer wishes to find the most probable utterance that could have caused the observed acoustic index string. That probability is the product of two factors: the probability that the utterance will produce the string and the probability that the speaker will wish to produce the utterance (the language model probability). Even (HMM) of speech production that, in principle, operates as follows: to each word v of vocabulary V, there corresponds an HMM of speech production. A concrete example of its structure is given in Fig. 1. The model of speech production of a sequence ofwords Wis a concatenation of models of individual words wi making up the sequence W (see Fig. 2).We recall that the HMM of Fig. 1 Fig. 1) and that they differ from each other in the number of states they have, in the values of the probabilities p(t) of the interstate transitions, and in the distributions q(a It) of generating output a when the (nonnull) transition t is taken.The first task, discussed in the second and third sections of this paper, is to show how the values of the parametersp(t) and q(a It) may be estimated from speech data, once the structure of the models (including the number of states) is selected by the system designer.Returning to formula 1, we are next interested in modeling P(W). We [5] (the trigram model). In this paper it is assumed that the equivalence classifier 4 was selected by the designer, who also estimated the language model probabilities P(v 4) for all words 9964 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

show abstract

A Maximum Likelihood Approach to Continuous Speech Recognition

Cited by 906 publications

References 12 publications

Acoustic and lexical resource constrained ASR using language-independent acoustic model and language-dependent probabilistic lexical model

Acoustic and lexical resource constrained ASR using language-independent acoustic model and language-dependent probabilistic lexical model

State of the art in continuous speech recognition.

Training and search methods for speech recognition.

Contact Info

Product

Resources

About