<?xml version="1.0" standalone="yes"?>
<Paper uid="H92-1037">
  <Title>Minimizing Speaker Variation Effects for Speaker-Independent Speech Recognition</Title>
  <Section position="5" start_page="193" end_page="194" type="concl">
    <SectionTitle>
5. SUMMARY
</SectionTitle>
    <Paragraph position="0"> In this paper, the codeword-dependent neural network (CDNN) was presented for speaker-independent speech recognition. The network was used as a nonlinear mapping function to transform speech data between speakers in each cluster and the golden speaker cluster. Performance evaluation showed that speaker-normalized front-end reduced the error rate by 15%, as shown in Figure 3, for the DARPA  tion. If we compare the error rate of speaker-dependent and speaker-independent systems, this 15 % error reduction is relatively small. We believe that the quality of mapping functions is extremely important if we want to bridge the gap between speaker-dependent and speaker-independent systems.</Paragraph>
  </Section>
class="xml-element"></Paper>