<?xml version="1.0" standalone="yes"?>
<Paper uid="W94-0208">
  <Title>SEGMENTING SPEECH WITHOUT A LEXICON: THE ROLES OF PHONOTACTICS AND SPEECH SOURCE</Title>
  <Section position="5" start_page="87" end_page="87" type="evalu">
    <SectionTitle>
RESULTS
</SectionTitle>
    <Paragraph position="0"> Each simulation was scored for the number of correct segmentation points inserted, as compared to the natural English segmentation. From this scoring, two values were computed: recall, the percent of all correct segmentation points that were actually found; and accuracy, the percent of the hypothesized segmentation points that were actually correct. In terms of hits, False alarms and misses, we have: hits recall : hits + misses hils accuracy : hits + false alarms Results are given in Table 1. Note that there is a trade-off between recall and accuracy--if all possible segmentation points were added, recall would be 100% but accuracy would be low; likewise, if only one segmentation point was added I)ctwccn two words, accuracy would be 100% but recall would be low. Since our goal is to correctly s(.gl|lelit speech, accuracy is more important th;m finding every correct segmentation. I&amp;quot;or exa.ml)h~, deciding 'littlekitty' is ;~ word is less disastrous than deciding 'li', 'tle', 'ki' and 'ty' are all words, because assigning meaning to 'littlekitty' is a reasonable first try at learning word-meaning pairs, whereas trying to assign separate meanings to 'li' and 'tle' is problematic.</Paragraph>
    <Paragraph position="1"> Tile i)erl~)rlnan(:e of i)IS'I'-I)IIONO Oll ddhl(lir('clcd Sl)oech shows l, hal, this systenl goes a long way toward solving the segnleutation i)rol)lcm.</Paragraph>
    <Paragraph position="2"> liowever, comparing the average pertbrmanees of simulations is also useful. The effect of phonetactic information can be seen by comparing the average performances of RAND-FREE and RAND-I'IIONO, since the C/)nly difference I)etw('(m l.h('m is the addit,ion of phonotactic constra, ints on segmentations in the, latl;er. Clearly I)houol, actic c(mstraints are useful, as both recall an(I accuracy improve. A similar comparison between RAND-IQtEE and DtST-FREE shows that distributional inlbrmation alone also improves performance. Note in all the results of D\[ST-FaEE that using distributional information alone favors recall over accuracy; in fact, the segmentation hypotheses produced by DIST-FREE have most words broken into single phoneme units with only a handful of words remaining intact. Two comparisons are nee(ted to show that the cond)ination of disl.rilmtional and phonotactic information I)erfi)rnm I)(,tter than either sour(:e a.lolle: DIST-PIIONO COml)ared to I{,ANI)-i~IIONO, to see tin'. elrect of a(Idiug disl.ributionai analysis to phonotactic constraints, and DIST-PIIONO compared to l)IsT-I,'ltl,:l,;, to see l.he effect of adding phonotactic constraints to distri-butional analysis. The former comparison shows that the sources corrlbined are more useful than phonotactic information alone, rl'he latter comparison is less obvious--the trade-off between recall and accuracy seems to have reversed, with no clear winner 5. Data on discovered word types helps make this conlparison: DIS'I'-I?IU,:I,~ found 12% of the words with 30% ac(:llraey an(1. Dis'r-PltONO found 33% of the words with 50% accuracy. Whereas the segmentation point data are inconclusive, word type data demonstrate that combining information sources is more usefifl than using distributional information alone.</Paragraph>
    <Paragraph position="3"> There is no obvious difference in performance between child- and adult-directed speech, except in DIST-PI1ONO (combined information sources) in which the difference is striking: accuracy remains high and recall rate more than tril)les for child-directed speeclL This difference is again supported by word type data: 14% recall with 30% accuracy for adult-directed speech, 56% recall with 65% accuracy for chihl-directed Sl)eech.</Paragraph>
  </Section>
class="xml-element"></Paper>