Pepper: Handling a multiverse of formats

<p>With the rising importance of empirical data in many fields of linguistic research, we see an<br> increase not only in the amount of electronically available corpora, but also in the number of<br> tools used to make this data accessible, processable and searchable. Most of these tools have<br> been developed in the course of specific linguistic projects and therefore can only handle a<br> certain kinds of linguistic information, such as syntactic-structures (e.g. TIGERSearch, Lezius<br> 2002), or dialogue-structures (e.g. EXMARaLDA, Schmidt 2004) etc. At the same time, each<br> tool uses its own, proprietary format for representing the text and its annotations. Such<br> formats are optimized for a specific kind of analysis and the performance of a specific<br> processing tool. Consequently they cannot easily be mapped onto each other. This impedes<br> those linguistic research questions which pre-suppose a global view on data, i.e., which<br> require the option to correlate, query and analyze several kinds of linguistic annotations at<br> once.<br> We present Pepper, a modularized converter framework addressing the problem that a<br> linguistic researcher may be limited to a small set of questions due to the tool(s) he or she<br> uses. Pepper is based on the meta-model Salt (Zipser & Romary 2010) and offers the<br> possibility of converting data from n formats into m formats with a minimal number of<br> necessary mappings. The pluggable architecture of Pepper allows the injection of new formats<br> into the framework. Pepper has no restrictions on the underlying techniques used in<br> representing these formats (e.g. XML, tabular-formats, bracketing-formats or mixtures<br> thereof).