Normalising Audio Transcriptions for Unwritten Languages

The task of documenting the world’s languages is a mainstream activity in linguistics which is yet to spill over into computational linguistics. We propose a new task of transcription normalisation as an algorithmic method for speeding up the process of transcribing audio sources, leading to text collections of usable quality. We report on the application of sentence and word alignment algorithms to this task, before describing a new algorithm. All of the algorithms are evaluated over synthetic datasets. Although the results are nuanced, the transcription normalisation task is suggested as an NLP contribution to the grand challenge of documenting the world’s languages.