Search

We present a brief overview of the CALLector project, and consider ethical questions arising from its overall goal of creating a social network to support creation and use of online CALL resources. We argue that these questions are best addressed in a decentralised, pluralistic open source architecture., 6 pages; based on talk presented at enetCollect WG3 & WG5 Meeting, Leiden, Holland, 2018

International audience

We describe Open CALL-SLT, a framework that allows people who may only have modest computer skills to create interactive speech-enabled CALL courses on the web. User content is uploaded to a server and remotely compiled and deployed. Courses can be created at six levels of increasing sophistication, where the specification of the content ranges from simple prompt/response pairs at one end to multimedia-enabled gamified dialogues with multiple paths at the other. We briefly describe the overall framework and the different levels.

We describe a Swedish version of CALL-SLT,a web-deployed CALL system that allows beginner/intermediate students to practise generativespoken language skills. Speech recognitionis grammar-based, with language modelsderived, using the Regulus platform, fromsubstantial domain-independent feature grammars.The paper focusses on the Swedishgrammar resources, which were developedby generalising the existing English featuregrammar into a shared grammar for Englishand Swedish. It turns out that this can be donevery economically: all but a handful of rulesand features are shared, and English grammaressentially ends up being treated as a reducedform of Swedish. We conclude by presentinga simple evaluation which compares theSwedish and French versions of CALL-SLT.

We present an evaluation of a Web-deployed spoken language CALL system, carried out using crowd-sourcing methods. The system, “Survival Japanese”, is a crash course in tourist Japanese implemented within the platform CALL-SLT. The evaluation was carried out over one week using the Amazon Mechanical Turk. Although we found a high proportion of attempted scammers, there was a core of 23 subjects who used the system in a responsible manner. The evidence that these subjects learned from their 111 sessions and 9092 spoken interactions was significant at P=0.001. Our conclusion is that crowd-sourcing is a potentially valid method for evaluating spoken CALL systems.

We describe a series of experiments in which we start with English to French and English to Japanese versions of an Open Source rule-based speech translation system for a medical domain, and bootstrap correspondign statistical systems. Comparative evaluation reveals that the rule-based systems are still significantly better than the statistical ones, despite the fact that considerable effort has been invested in tuning both the recognition and translation components; also, a hybrid system only marginally improved recall at the cost of a los in precision. The result suggests that rule-based architectures may still be preferable to statistical ones for safety-critical speech translation tasks. Describimos una serie de experimentos en los que comenzamos con las versiones de inglés a francés y de inglés a japonés de un sistema de traducción oral basado en reglas de código abierto para un dominio médico, e iniciamos los sistemas estadísticos correspondientes. La evaluación comparativa revela que los sistemas basados en reglas son significativamente mejores que los estadísticos, a pesar del hecho de que se ha dedicado un esfuerzo considerable en mejorar los componentes de reconocimiento y traducción; además, un sistema híbrido sólo mejoró de modo marginal la memoria a costa de una pérdida de precisión. Los resultados indican que las arquitecturas basadas en reglas podrían ser mejores para tareas de traducción oral críticas. Descrivim una sèrie d'experiments en què comencem amb les versions d'anglès a francès i d'anglès a japonès d'un sistema de traducció oral basat en regles en codi obert per a un domini mèdic, i iniciem els sistemes estadístics corresponents. L'avaluació comparativa revela que els sistemes basats en regles encara són significativament millors que els estadístics, malgrat el fet que s'ha dedicat un esforç important a millorar els components tant de reconeixement com de traducció; a més a més, un sistema híbrid només va millorar marginalment la memòria a costa d'una pèrdua de precisió. Els resultats indiquen que les arquitectures basades en regles podrien ser millors que les estadístiques per a tasques de traducció oral crítiques.

We argue that there is an urgent need for spoken dialogue translation systems in the medical domain. In this work we describe how an existing system of this kind, originally intended for a desktop PC, was adapted to run in a mobile environment. The different characteristics of the target device necessitate new interaction approaches and interface design. The new configuration uses two client applications running on two different devices, one for the doctor and one for the patient, together with a server; the components are connected over a wireless network. The system supports context-dependent translation in both directions. We give a general overview of the system, and discuss some of the relevant design issues pertaining to deployment of speech-enabled systems on mobile devices.

We describe recent work on MedSLT, a medium-vocabulary interlingua-based medical speech translation system, focussing on issues that arise when handling languages of which the grammar engineer has little or no knowledge. We describe how we can systematically create and maintain multiple forms of grammars, lexica and interlingual representations, with some versions being used by language informants, and some by grammar engineers. In particular, we describe the advantages of structuring the interlingua definition as a simple semantic grammar, which includes a human-readable surface form. We show how this allows us to rationalise the process of evaluating translations between languages lacking common speakers. The grammar-based interlingua definition can also be used in other ways. We describe two applications: a simple generic tool for debugging to-interlingua translation rules, and a method for improving speech understanding performance by rescoring N-best speech hypothesis lists. Examples presented focus on the concrete case of translation between Japanese and Arabic in both directions.

Particularly considering the requirement of high reliability, we argue that the most appropriate architecture for a medical speech translator that can be realised using today's technology combines unidirectional (doctor to patient) translation, medium-vocabulary controlled language coverage, interlingua-based translation, an embedded help component, and deployability on a hand-held hardware platform. We present an overview of the Open Source MedSLT prototype, which has been developed in accordance with these design principles. The system is implemented on top of the Regulus and Nuance 8.5 platforms, translates patient examination questions for all language pairs in the set {English, French, Japanese, Arabic, Catalan}, using vocabularies of about 400 to 1 100 words, and can be run in a distributed client/server environment, where the client application is hosted on a Nokia Internet Tablet device.

We present an overview of MedSLT, a medium vocabulary medical speech translation system, focussing on the representational issues that arise when translating temporal and causal concepts. Although flat key/value structures are strongly preferred as semantic representations in speech understanding systems, we argue that it is infeasible to handle the necessary range of concepts using only flat structures. By exploiting the specific nature of the task, we show that it is possible to implement a solution which only slightly extends the representational complexity of the semantic representation language, by permitting an optional single nested level representing a subordinate clause construct. We sketch our solutions to the key problems of producing minimally nested representations using phrase-spotting methods, and writing cleanly structured rule-sets that map temporal and phrasal representations into a canonical interlingual form.

We present an overview of MedSLT, an Open Source platform for developing limited-domain medical speech translation systems. We focus in particular on the speech understanding architecture, which uses grammar-based language models derived using corpus-based specialisation methods from a single linguistically motivated grammar, and summarise the results of two evaluations which investigate the appropriateness of these design choices. Other sections describe the interlingua and its relationship with the recognition architecture, and the current demo system.

The paper directly compares two versions of a medical speech translation system, one with a grammar based language model (GLM) recognizer and the other with a statistical language model (SLM) recognizer. We construct the GLM using a corpus-based method, so that both the GLM and the SLM can be derived from the same corpus; evaluation is carried out with respect to performance on the speech translation task. Despite using a very small training set for both the GLM and the SLM, the SLM delivers much better word error rates on unseen test material. Nonetheless, evaluating both systems on translation performance rather than word error rates, the GLM-based version of the system outperforms the SLM on the actual translation task.

We describe an architecture for implementing spoken natural language dialogue interfaces to semi-autonomous systems, in which the central idea is to transform the input speech signal through successive levels of representation corresponding roughly to linguistic knowledge, dialogue knowledge, and domain knowledge. The final representation is an executable program in a simple scripting language equivalent to a subset of Cshell. At each stage of the translation process, an input is transformed into an output, producing as a byproduct a "meta-output" which describes the nature of the transformation performed. We show how consistent use of the output/meta-output distinction permits a simple and perspicuous treatment of apparently diverse topics including resolution of pronouns, correction of user misconceptions, and optimization of scripts. The methods described have been concretely realized in a prototype speech interface to a simulation of the Personal Satellite Assistant., Working notes from AAAI Spring Symposium

Speech is becoming increasingly popular as an interface modality, especially in hands- and eyes-busy situations where the use of a keyboard or mouse is difficult. However, despite the fact that many have hailed speech as being inherently usable (since everyone already knows how to talk), most users of speech input are left feeling disappointed by the quality of the interaction. Clearly, there is much work to be done on the design of usable spoken interfaces. We believe that there are two major problems in the design of speech interfaces, namely, (a) the people who are currently working on the design of speech interfaces are, for the most part, not interface designers and therefore do not have as much experience with usability issues as we in the CHI community do, and (b) speech, as an interface modality, has vastly different properties than other modalities, and therefore requires different usability measures., Position paper for CHI 2000 Workshop on Natural-Language Interaction

We describe an implemented spoken-language dialogue system for a travel-planning domain, which accesses a commercially available travel-information web-server and supports a flexible mixed-initiative dialogue strategy. We argue, based on data from initial Wizard-of-Oz experiments, that mixed-initiative strategies are appropriate for many types of user, but require more sophisticated architectures for processing of language and dialogue; we then use these observations to motivate an architecture which combines parallel deep and shallow natural language analysis engines and an agenda-driven dialogue manager. We outline the top-level processing strategy used by the dialogue manager, and also a novel formalism, which we call Flat Utterance Description, that allows us to reduce the output of the deep and shallow language-processing engines to a common representation.

We describe two methods relevant to multi-lingual machine translation systems, which can be used to port linguistic data (grammars, lexicons and transfer rules) between systems used for processing related languages. The methods are fully implemented within the Spoken Language Translator system, and were used to create versions of the system for two new language pairs using only a month of expert effort., 6 pages, needs aclap.sty. To appear in "From Research to Commercial Applications" workshop at ACL-97, see also http://www.cam.sri.com

The thesis describes a logical formalization of natural-language database interfacing. We assume the existence of a ``natural language engine'' capable of mediating between surface linguistic string and their representations as ``literal'' logical forms: the focus of interest will be the question of relating ``literal'' logical forms to representations in terms of primitives meaningful to the underlying database engine. We begin by describing the nature of the problem, and show how a variety of interface functionalities can be considered as instances of a type of formal inference task which we call ``Abductive Equivalential Translation'' (AET); functionalities which can be reduced to this form include answering questions, responding to commands, reasoning about the completeness of answers, answering meta-questions of type ``Do you know...'', and generating assertions and questions. In each case, a ``linguistic domain theory'' (LDT) $\Gamma$ and an input formula $F$ are given, and the goal is to construct a formula with certain properties which is equivalent to $F$, given $\Gamma$ and a set of permitted assumptions. If the LDT is of a certain specified type, whose formulas are either conditional equivalences or Horn-clauses, we show that the AET problem can be reduced to a goal-directed inference method. We present an abstract description of this method, and sketch its realization in Prolog. The relationship between AET and several problems previously discussed in the literature is discussed. In particular, we show how AET can provide a simple and elegant solution to the so-called ``Doctor on Board'' problem, and in effect allows a ``relativization'' of the Closed World Assumption. The ideas in the thesis have all been implemented concretely within the SRI CLARE project, using a real projects and payments database. The LDT for the example database is described in detail, and examples of the types of functionality that can be achieved within the example domain are presented., Comment: 162 pages, Latex source, PhD thesis (U Stockholm, 1993). Uses style-file ustockholm_thesis.sty

The paper describes a Swedish-language customization (S-CLE) of the SRI Core Language Engine, which has been developed at SICS from the original English-language version by replacing English-specific modules with corresponding Swedish-language versions. The S-CLE is intended to be used as a building block in a broad range of applications, such as data-base query system, machine translation systems, NL front-ends, speech-to-text/text-to-speech systems, and so on. Examples of the first two types of application already exist. The main part of the S-CLE is an extensive Swedish grammar that is compiled into parsing and generation modules. The grammar formalism is a type of unification grammar loosely based on Generalized Phrase Structure Grammar (GPSG). Generation is performed using the Semantic-Head-Driven algorithm. Analysis turns sentences into ``Quasi-Logical Form'' (QLF), a logical-form representation, while generation works in the opposite direction. Intermediate stages include processing of morphology, syntax and semantics. For knowledge-base applications, a separate module can convert QLFs into conventional scoped logical forms. After two-and-a-half years of work (approximately 45 person months), the first prototype system has a vocabulary of about 1900 words and covers a fairly broad range of possible grammatical constructions. Based on our experience in this project, we present in this paper detailed arguments to support the claim that customization of an English-language NLP system is a highly cost-effective way of constructing Swedish language systems with corresponding functionality. A shorter version of this paper appears in L. Ahrenberg (ed.): Papers from the Third Nordic Conference on Text Comprehension in Man and Machine , Link Sweden, 1992.

The syntactic analysis component of a large-scale natural language query interface to relational data bases was greatly sped up by applying explanation-based learning(EBL), a machine learning technique. The idea is that one for most input queries can bypass the ordinary parser, instead using a set of learned rules. When no rule applies, one must pay the price of a small overhead. The set of learned rules is extracted automatically from sample queries posed by a user, thus tuning the system for that particular user. Several non-trivial problems, arising from the characteristics of the target system, were solved during the project. Measurements on a small test corpus indicated that the speed-ups when a learned rule could be used are on average a factor 30 and that the overhead, when no rule applied, is less than 3 percent. Extended version of a paper titled "Using Explanation-Based Learning to Increase Performance in a Large-Scale NL Query System" by the same authors, that was presented at the third DARPA Workshop on Speech and Natural Language, Hidden Valley, 1991. Original report number R91001.

This document is the final report for a research project aimed at producing a prototype system for on-line translation of typed dialogues between speakers of different natural languages. The work was carried out jointly by SICS and SRI Cambridge. The resulting prototype system (called Billingual Conversation Interpreter, or BCI) translates between english and Swedish in both directions.The Major components of the BCI are two copies of the SRI Core Language Engine, equipped with English and Swedish grammars respectively. These are linked by the transfer and disambiguation components. Translation takes place by analyzing the source-language sentence into Quasi Logical Form ( QLF), a linguistically motivated logical representation, transferring this into a target-language QLF, and generating a target-language sentence. When ambiguities occur that cannot be resolved automatically, they are clarified by Querying the appropriate user. The clarification dialogue presupposes no knowledge of either linguistics or the other language. The prototype system has a broad grammatical coverage, a initial vocabulary of about 1000 words together with vocabulary expansion tools, and a set of English-Swedish transfer rules. The formalism developed for coding this linguistic information make it relatively easy to extend the system. We believe that the project was successful in demonstrating the feasibility of using these techniques for interactive translation applications, and provides a sound basis for development of a large scale message translator system with potential for commercial exploitation.The main sections of this report are the following: * A non-technical introduction, summarizing the BCI's design , and containing a sample session. * An overview of the Swedish version of the CLE. * A detailed discussion of the theory and practice of QLF transfer. * A description of the interactive disambiguation method. * Suggestions for possible follow-on projects aimed in the direction of practically usable commercial systems. This is a joint report by SICS and SRI, which is publ. by both institutes. SRI Cambridge Report CCSRC-018.

This Document is an introduction to a research project aimed at producing a prototype system for on-line translation of typed dialogues between speakers of different natural languages. The work was carried out jointly by SICS and SRI Cambridge. The resulting prototype system (called Billingual Conversation Interpreter, or BCI) translates between English and Swedish in both Directions. The major components of the BCI are two copies of the SRI Core Language Engine, equipped with English and Swedish grammars respectively. These are linked by the transfer and disambiguation components. Translation takes place by analyzing the source-language sentence into Quasi Logical Form (QLF), a linguistically motivated logical representation, transferring this onto a target-language QLF, and generating a target-language sentence. We believe that the project was successful in demonstrating the feasibility of using these techniques for interactive translation applications, and provides a sound basis for development of a large scale message translator system. The final section of the paper points to several possible follow-on projects aimed in the direction of practically usable commercial systems.

We argue that bidding in the game of Contract Bridge can profitably be regarded as a micro-world suitable for experimenting with pragmatics. We sketch an analysis in which a "bidding system" is treated as the semantics of an artificial language, and show how this "language", despite its apparent simplicity, is capable of supporting a wide variety of common speech acts parallel to those in natural languages; we also argue that the reason for the relatively unsuccessful nature of previous attempts to write strong Bridge playing programs has been their failure to address the need to reason explicitly about knowledge, pragmatics, probabilities and plans. We give an overview of Pragma, a system currently under development at SICS, which embodies these ideas in concrete form, using a combination of rule-based inference, stochastic simulation, and "neural-net" learning. Examples are given illustrating the functionality of the system in its current form. Original report number R90011.

Work by Yoav Shoham advocating the use of default logic is critically examined, and several serious inconsistencies are pointed out. Shoham's analysis of the frame problem is shown to depend on a misinterpretation of the formal basis for Newtonian mechanics, and strong reasons are given to doubt the correctness of his arguments for reducing auto-epistemic reasoning to default logic. Detailed counter-examples are provided for both cases. Original report number R88009. Related paper "Did Newton Solve the Extended Prediction problem?" by M. Rayner published in Proceedings 1st Intl. Conf. on the Foundations of Knowledge Representation, R. Brachman & H. Levesque, eds., Morgan Kaufman, 1989.

We present a framework for reasoning about the concepts of "knowing what" and "finding out", in which the key concept is to identify "finding out the answer to question Q" with "achieving a situation in which Q is decidable" . We give examples of how the framework can be used to formulate non-trivial problems involving the construction of plans to acquire and use information, and go on to demonstrate that these problems can often be solved by systematic application of a small set of goal-directed backward-chaining rules. In conclusion, it is suggested that systems of this kind are potentially implementable in l-Prolog, a logic programming language based on higher-order logic. Original report number R89016.

We demonstrate how to reason about plans by treating them as programs expressed in a simple imperative language, and illustrate this using an example originally suggested by McCarthy and later discussed at length by McDermott. A detailed correctness proof is provided, which it is argued is greatly superior to McDermott's. Original report number R88003. Related paper, "Reasoning about Plans Represented as Procedural Programs" located in Artificial Intelligence III: Methodology, Systems, Applications; Proceedings of the Third International Conference on Artificial Intelligence: Methodology, Systems, Applications (AIMSA '88), Varna, Bulgaria, 1988, pp. 93-102. T. O'Shea and V. Sgurev, editors, North-Holland Press.

We describe a fairly comprehensive handling of the syntax and semantics of comparative constructions. The analysis is largely based on the theory developed by Pinkham, but we advance arguments to support a different handling of phrasal comparatives - in particular, we replace the use of C-ellipsis with a method of interpretation we call contrastive comparison. We explain the reasons for dividing comparative sentences into different categories, and for each category we give an example of the corresponding Montague semantics. The ideas have all been implemented within a large-scale grammar for Swedish, a "toy" version of which is presented, along with examples of the output. This report is a revised and extended version of two earlier papers: (1) "Comparatives in Logic Grammars-Two Viewpoints" located in Natural Language Understanding and Logic Programming II: Proceedings of the 2nd Int'l Workshop on Natural Language Understanding and Logic Programming, Vancouver, Canada, 1987, pp. 153-168, V. Dahl and P. Saint-Dizier, editors, North-Holland Press, 1988; and (2) "Parsing and Interpreting Comparatives" located in Proceedings of the 26th Annual Conference of the Ass. for Computational Linguistics (ACL), Buffalo, 1988, published by the ACL, 1988. Original report number R88018.

The paper criticizes arguments recently advanced by Shoham, McDermott and Sandewall, which purport to demonstrate the relevance of non-monotonic logic to the formalization of reasoning about the evolution of mechanical systems in continuous time. The first half of the paper examines the "Extended Prediction Problem" of Shoham and McDermott; reasons are given to support the claim that the "problem" is the product of a mistaken understanding of the the formal basis of Newtonian mechanics, and has no real existence. An example is given showing how, contrary to Shoham and McDermott's arguments, it is possible to formalise reasoning about the evolution of physical systems in continuous time using only classical logic and differential calculus. The second half then reviews Sandewall's non-monotonic logic for almost-continuous systems. Here it is argued that the proposed framework offers only very marginal advantages in compactness of notation, and generally tends to collapse back into classical logic. In summary, I conclude that there is as yet no good reason to believe that non-monotonic logic will be a useful tool in this area. Original report number R98013.

It is shown how ideas adapted from recent work on explanation-based generalization can be used to allow a logic grammar to "learn" useful derived grammar rules by generalizing them from example sentences. The method is presented in the form of a small Prolog meta-interpreter, and its soundness is formally proved. Examples are given showing the application of the generalizer, first to a toy grammar with 40 rules and then to a largish independantly developed system which involves non-trivial syntactic and semantic analysis. Original report number R89014.

Explanation-based learning is a technique which attempts to optimize performance of a rule-based system by adding new rules constructed from generalizations of successfully-solved examples. The paper summarizes previous work showing how this idea can be used in natural language processing, and describes experiments in which the EBL method was applied to the CHAT-80 system of Pereira and Warren. In particular, we address the problem of assuring the utility of learning a rule, since the benefit of a learned rule may not outweigh the increased search time incurred in checking its applicability. We show that this problem can be overcome in the NL domain by indexing acquired rules by their lexical constraints, which in general vastly reduces the number of potentially applicable rules. Such an indexing method was implemented and timing studies were made comparing its access speed to that of normal linear search. The indexing scheme required an average access time of 35 - 40 ms independent of the number of learned rules. The results suggest that the overhead of the indexing scheme is small.