The Hearsay-II Speech-Understanding System: Integrating Knowledge to Resolve Uncertainty

  title={The Hearsay-II Speech-Understanding System: Integrating Knowledge to Resolve Uncertainty},
  author={Lee D. Erman and Frederick Hayes-Roth and Victor R. Lesser and Raj Reddy},
  journal={ACM Comput. Surv.},
The Hearsay-II system, developed during the DARPA-sponsored five-year speech-understanding research program, represents both a specific solution to the speech-understanding problem and a general framework for coordinating independent processes to achieve cooperative problem-solving behavior. As a computational problem, speech understanding reflects a large number of intrinsically interesting issues. Spoken sounds are achieved by a long chain of successive transformations, from intentions… 
HEARSAY-III: A Domain-Independent Framework for Ex
This paper describes the Hearsay-Ill framework, a conceptually simple extenslon of the basic ideas in the Hearsey-II speech-understanding system, concentrating on its departures from Hearsy-II.
Linguistic Processing in a Speech Understanding System
The goal of a speech understanding system is to correctly identify the action to be taken as a response to a user’s voiced request, and linguistic constraints are integrated into the recognizer, which decodes one string of words that is treated by a natural language interface.
The challenge of spoken language systems: research directions for the nineties
The need for multidisciplinary research is reviewed, for development of shared corpora and related resources, for computational support and far rapid communication among researchers, and the expected benefits of this technology are reviewed.
Learning Fault-Tolerant Speech Parsing with SCREEN
The goal for this approach is to explore the parallel interactions between various knowledge sources for learning incremental fault-tolerant speech parsing in a system SCREEN using various hybrid connectionist techniques.
A multiknowledge base system for continuous speech understanding
  • Y. Gong, J. Haton
  • Physics
    [1990] Proceedings. 10th International Conference on Pattern Recognition
  • 1990
A spoken Chinese understanding system which accepts continuous speech and produces Lisp-executable functions is presented. A multiple-knowledge-source organization model, syntactic structure
Extraction of acoustic cues using a grammar of frames
  • R. Mori
  • Computer Science
    Speech Commun.
  • 1983
Real-time linguistic analysis for continuous speech understanding
This paper describes the approach followed in the development of the linguistic processor of the continuous speech dialog system implemented at the labs, and results are discussed, as obtained from an implementation of the system on a Sun SparcStation 1 using the C language.
HEARSAY-II: A Domain-Independent Framework for Expert Systems
This paper describes the Hearsay-III framework, concentrating on its departures from Hearsey-II, and describes the use of production systems to encapsulate expert knowledge in manageable and relatively independent chunks.
Second Thoughts on an Artificial Intelligence Approach to Speech Understanding
This paper analyzes the roots of this failure as a case study in AI methodology gone awry and explains why the original, classicly AI goals — namely, be optimal in principle, be well integrated, iteratively refine the interpretation, deal directly with noisy inputs, be linguistically interesting, be tunable by hand, work with clear hypotheses, and relate to general issues in AI — are less important than they seemed.


The Hearsay‐II System has as its design goal recognition, understanding, and responding to connected speech utterances, particularly in situations where sentences cannot be guaranteed to agree with
Organization of the Hearsay II speech understanding system
The issues of the system organization of the HSII system are dealt with, which include a convenient modular structure for incorporating new knowledge into the system at any level, and a system structure suitable for execution on a parallel processing system.
A Retrospective View of the Hearsay-II Architecture
Experiences gained while successfully applying the Hearsay-II architecture to the problem of speech understanding are described, and the paradigm of viewing problem solving in terms of hypothesize-and-test actions distributed among distinct representations of the problem has been shown to be computationally feasible.
The HARPY speech recognition system
The HARPY system is the result of an attempt to understand the relative importance of various design choices of two earlier speech recognition systems developed at Carnegie-Mellon University, in which knowledge is represented as a finite state transition network but without the a-priori transition probabilities.
The Hearsay-I Speech Understanding System: An Example of the Recognition Process
The structure and operation of the Hearsay-I1speech understanding system is described by considering its operation in a particular task situation: Voice-Chess and preliminary results of the reduction in search resulting from the use of various sources of knowledge are given.
Word hypothesization for large-vocabulary speech understanding systems.
The thesis presents the design and performance of a bottom-up word hypothesizer (Noah) capable of handling very large vocabularies and suggests that speech understanding systems for general English can obtain a great amount of constraint from the acoustics alone.
Harpy, production systems and human cognition
A viable and interesting theory of human speech perception has been generated by constructing a psychological model of speech perception that is faithful to Harpy and asking whether it is acceptable given what the authors know about human processing capabilities.
Selection of word islands in the Hearsay-II speech understanding system
In Hearsay-II, a word recognizer hypothesizes words bottom-up from acoustic data to form multi-word islands which the syntax-level knowledge source first checks for grammatically and then attempts to extend to form a complete recognition.
Speech Understanding Systems
Noah-A Bottom-Up Word Hypothesizer for Large-Vocabulary Speech Understanding Systems
  • A. Smith, L. Erman
  • Linguistics
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 1981
The design and performance of Noah are presented, a bottom-up word hypothesizer which is capable of handling large vocabularies-more than 10 000 words, and the performance is presented as a function of the vocabulary size.