The Speech Accent Archive: towards a typology of English accents

The theoretical and practical value of studying human accented speech is of interest to linguists, language teachers, speech recognition researchers, and computational linguists. It is also part of the research program behind the Speech Accent Archive. The Archive is a growing annotated corpus of English speech varieties. All samples include a complete digital audio version and a narrow phonetic transcription. Each speaker is located geographically and crucial demographic parameters are supplied. This paper will discuss the architecture of the Speech Accent Archive and a computational mechanism we developed to assist in analyzing the archive, the Speech Transcription Analysis Tool (STAT). We show how such a tool generates a “typological corpus” of English accents that can benefit researchers in the fields of theoretical and applied linguistics.