File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/93/e93-1039_concl.xml
Size: 2,645 bytes
Last Modified: 2025-10-06 13:56:56
<?xml version="1.0" standalone="yes"?> <Paper uid="E93-1039"> <Title>apos;Structure and Usage of Current Bell Labs Text</Title> <Section position="10" start_page="337" end_page="338" type="concl"> <SectionTitle> 7 Results </SectionTitle> <Paragraph position="0"> The system just described produces sharp and natural-sounding distinctions of intonation contour in minimal pairs of queries like the following: 14The ~ symbol separates syntactic categories from their corresponding prosodic categories and lexical items from their pitch/boundary markings.</Paragraph> <Paragraph position="1"> Examples (29) and (30) illustrate the ability of our system to produce appropriately different intonation contours for identical strings of words depending on the context, which determines the information structure of the response..If the responses in these examples are interchanged, the result sounds distinctly unnatural in the given contexts. From examples (31) and (32), it will be apparent that our system has the ability to make distinctions in focus placement within themes and rhemes based on context. The issue of focus placement can be crucial in more complex themes and rhemes, as shown below: (33) Q: I know the old widget has the slowest processor, but which widget has the FASTEST processor? As noted earlier, such precisely specified themes are uncommon in normal dialogue. Consequently, the Null Tone Promotion rule is employed for unmarked themes, allowing the types of responses in (36) and (37) below. The theme is taken to be the longest possible prosodically unmarked constituent allowed by the syntax.</Paragraph> <Paragraph position="2"> (36) Q: I know that programmers use widgets, but which people DESIGN widgets? Although we have only briefly discussed the possibility of multiple pitch accents within a theme or rheme, we have included such a capability in our implementation. The system's ability to handle multiple pitch accents is illustrated by the following example. null (38) Q: I know that students USE WODGETS, but which people DESIGN WIDGETS? H* H* LL% A: ENGINEERS design widgets.</Paragraph> <Paragraph position="3"> H* L While many important problems remain, examples like these show that it is possible to produce synthesized speech with contextually appropriate intonational contours using a combinatory theory of prosody and information structure that is completely transparent to syntax and semantics. The model of utterance generation for Combinatory Categorial Grammars presented here implements the prosodic theory in a similarly transparent and straightforward manner.</Paragraph> </Section> class="xml-element"></Paper>