<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-2903">
  <Title>Audio Hot Spotting and Retrieval Using Multiple Features</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
7. Conclusion
</SectionTitle>
    <Paragraph position="0"> In this paper, we have shown that by automatically detecting multiple audio features and making use of these features in a relational database, our Audio Hot Spotting prototype allows a user to begin to apply the range of cues available in audio to the task of multi-media information retrieval. Areas of interest can be specified using keywords, phrases, speaker identity, prosodic features, and information-bearing background sounds, such as applause and laughter. When matches are found, the system displays the recognized text and allows the user to play the audio or video in the vicinity of the identified &amp;quot;hot spot&amp;quot;. With the advance of component technologies such as automatic speech recognition, speaker identification, and prosodic and audio feature extraction, there will be a wider array of audio features for the multimedia information systems to query and retrieve, allowing the user to access the exact information desired rapidly.</Paragraph>
  </Section>
class="xml-element"></Paper>