<?xml version="1.0" standalone="yes"?>
<Paper uid="H92-1120">
  <Title>A Real-Time Spoken-Language System for Interactive Problem Solving</Title>
  <Section position="1" start_page="0" end_page="0" type="metho">
    <SectionTitle>
PROJECT GOALS
</SectionTitle>
    <Paragraph position="0"> The goal of this project, to develop a spoken-language interface to the Official Airline Guide database, is being developed along two overlapping research and development lines: one focussed on an SLS kernel for database query, and the other on the full interactive system.</Paragraph>
  </Section>
  <Section position="2" start_page="0" end_page="0" type="metho">
    <SectionTitle>
RECENT RESULTS
</SectionTitle>
    <Paragraph position="0"> * Evaluation of SRI's NL, SLS, and speech recognition technologies. SRI's February-92 weighted error rate for ATIS class A and D utterances combined was 31.0% for NL and 45.4% for SLS. SRI's February-92 SPREC word error rate for all ATIS utterances was 11.0%. For the speaker-independent, closed 5K vocabulary CSR task, SRI achieved a 16.6% word error rate in the verbalized punctuation test and a 17.1% word error rate in the non-verbalized-punctuation test, using the standard bigram language model.</Paragraph>
    <Paragraph position="1"> * Encouraging initial results in a study of verbal repairs in the ATIS MADCOW corpus; the study included analysis and classification of the repairs in the corpus as well as a method using both pattern matching techniques and acoustic evidence to detect and correct the repairs.</Paragraph>
    <Paragraph position="2"> * Implementation of RASTA filtering (high-pass filtering in the Log Spectral domain to remove timeinvariant or slow moving linear channel effects) to improve channel robustness.</Paragraph>
    <Paragraph position="3"> * Experiments showing differences between spontaneous and read speech and the need for including spontaneous speech in trainir/g, development, and test sets.</Paragraph>
    <Paragraph position="4"> * Development of infrastructure necessary for dealing with very large vocabularies and language models such as those associated with the CSR corpus.</Paragraph>
    <Paragraph position="5"> * Improvement in GEMINI natural-language processing system's linguistic coverage of a 2100-utterance ATIS traning corpus to 92% syntactic coverage, 77% semantic coverage.</Paragraph>
    <Paragraph position="6"> * Development of algorithms for tracking evolving discourse structure. On the development set of the ATIS corpus, we find these algorithms correctly determine the discourse context with greater than 90% accuracy.</Paragraph>
    <Paragraph position="7"> * Improvements to the template matcher: new templates, improved portability, more linguistic knowledge incorporated, reduced overgeneration, improved context handling, improved coverage of databas'e retrieval component, all without significantly increasing response time, which is still less than a second.</Paragraph>
    <Paragraph position="8"> * Analysis of data on human-machine problem solving using our SLS ATIS system: we have analyzed the effect on user satisfaction and system performance of system errors, user experience, and instructions to users. We have also explored trade-offs of time vs.speed and user satisfaction.</Paragraph>
    <Paragraph position="9"> * Collection of data (speech, transcriptions, and logfiles) using our SLS ATIS system. To date this includes 72 speakers, 2301 utterances, and 134 scenarios. null * Improvements in the SLS ATIS demo system, including better paraphrasing of systemn's understanding, easier to read displays, improved system error messages, simpler control of context mechanism.</Paragraph>
    <Paragraph position="10"> * Development with MIT of a method for end-to-end evaluation that takes into account the whole interaction; implementation of the method is in progress.</Paragraph>
  </Section>
  <Section position="3" start_page="0" end_page="487" type="metho">
    <SectionTitle>
PLANS FOR THE COMING YEAR
</SectionTitle>
    <Paragraph position="0"> * Fully integrate GEMINI into SRI's ATIS system.</Paragraph>
    <Paragraph position="1"> * Explore tighter integration of speech and NL processing for better overall SLS performance.</Paragraph>
    <Paragraph position="2"> * Improve speech recognition computational and word-accuracy performance for the ATIS and CSR domains with better models of the channel, speaker, and spontaneous speech effects.</Paragraph>
  </Section>
class="xml-element"></Paper>