<?xml version="1.0" standalone="yes"?>
<Paper uid="H94-1116">
  <Title>CONSISTENCY MODELING</Title>
  <Section position="1" start_page="0" end_page="0" type="metho">
    <SectionTitle>
CONSISTENCY MODELING
</SectionTitle>
    <Paragraph position="0"/>
  </Section>
  <Section position="2" start_page="0" end_page="0" type="metho">
    <SectionTitle>
PROJECT GOALS
</SectionTitle>
    <Paragraph position="0"> SKI's consistency modeling project beg~ in August 1992, The goal of the project is to develop consistency modeling technology. That is, we aim to reduce the number of improper independence assumptions used in traditional speech recognition algorithms so that the resulting speech recognition hypotheses are more selfconsistent and, therefore, more accurate. Consistency is achieved by conditioning HMM output distributions on state and observations histories, P(x\[s,H). The goal of the project is finding the proper form of the probability distribution P, the proper history vector, H, and the proper feature vector, x, and developing the infrastructure (e.g.</Paragraph>
    <Paragraph position="1"> efficient estimation and search techniques) so that consistency modeling can be effectively used.</Paragraph>
  </Section>
  <Section position="3" start_page="0" end_page="473" type="metho">
    <SectionTitle>
RECENT RESULTS
</SectionTitle>
    <Paragraph position="0"> Highlights of our accomplishments to date include a large reduction in our speech recognition error rate due to the development on Genonic HMM technology, and the development of a real-time version of our system. A summary of our accomplishments include: * SKI developed and refined Genonlc HMM technology. This is a form of continuous density I-IMM that, combined with other advances, allowed us to reduce our error rate by over a factor of two over the past year. Currently our best performance is 9.3% word error as measured on ARPA's 20K Nov.</Paragraph>
    <Paragraph position="1"> 1992 evaluation set and 13.6% on ARPA's 20K Nov.</Paragraph>
    <Paragraph position="2"> 1993 test set, both using ARPA's standard grammars.</Paragraph>
    <Paragraph position="3"> * SRI developed an information-theoretic framework for estimating the effect of the history H in the conditional HMM output distribution P(x/s,H) when H is constrained to be a previous frames xt_ i.</Paragraph>
    <Paragraph position="4"> * SRI implemented a version of continuous localconsistency modeling. SRI verified that the information-theoretic framework above indeed predicts recognition accuracy improvements. We  measured performance of continuous local consistency for several different frame lags.</Paragraph>
    <Paragraph position="5"> deg SKI developed progressive search: a framework for using hierarchies of recognition algorithm~ ill order to achieve fast yet accurate speech recognition.</Paragraph>
    <Paragraph position="6"> * SKI developed tree-based search schemes for implementing large-vocabulary speech recognition systems. This resulted in a real-time 20,000 recognition system with about 30% word error.</Paragraph>
    <Paragraph position="7"> * SKI developed Gaussian shortlist technology and other techniques for avoiding Gaussian distribution evaluation. This resulted in a net deecrease of Ganssian evaluations by a factor of 30 with no recognition accuracy degradation. By combining this with the above search technologies, we expect implement a real-time near full-accuracy speech recognition in the near future.</Paragraph>
    <Paragraph position="8"> * SKI tran~erred, improved, and evaluated feature mapping technology developed on NSF funding.</Paragraph>
    <Paragraph position="9"> This allows our system to be virtually microphone independent for large classes of microphones. For instnnce, using models developed for a Sennheiser close-talking microphone, accuracy degraded only 10% (5.9% to 6.4% error) when tested on data recorded with an Audio Technica desk-mounted microphone.</Paragraph>
  </Section>
class="xml-element"></Paper>