Towards a query language for annotation graphs

TL;DR

This paper introduces a specialized query language for annotation graphs, addressing the unique challenges of representing and querying complex speech database structures with temporal and hierarchical relationships.

Contribution

It develops a novel query language tailored for annotation graphs and presents optimization techniques for efficient querying on a relational model.

Findings

Designed a query language suitable for acyclic, unrooted annotation graphs

Proposed optimization strategies for relational query processing

Demonstrated applicability to speech database queries

Abstract

The multidimensional, heterogeneous, and temporal nature of speech databases raises interesting challenges for representation and query. Recently, annotation graphs have been proposed as a general-purpose representational framework for speech databases. Typical queries on annotation graphs require path expressions similar to those used in semistructured query languages. However, the underlying model is rather different from the customary graph models for semistructured data: the graph is acyclic and unrooted, and both temporal and inclusion relationships are important. We develop a query language and describe optimization techniques for an underlying relational representation.