Transcription as speech-to-text data transformation

This dissertation proposes a method of preparing natural language data for analysis, focusing on the transformation of spoken language into written form in a Language Variation framework. The protocol devised here is one that strives to maintain the integrity of the primary linguistic data under investigation, while providing a theoretically straightforward method of data preparation that places a reasonable time and resource constraint on the analyst. This project also seeks to establish practical guidelines for the transformation linguistic data from sound to transcript that will encourage and support data-sharing amongst the linguistic community and greater research community at large.