This dataset contains the first Indonesian speech dataset for structurally ambiguous utterances and each of transcription and two disambiguation texts.
The structurally ambiguous sentences were adapted from Types 4,5,6, and 10 of Types Of Syntactic Ambiguity in English by [Taha et al., 1983].
For each chosen type, 100 structurally ambiguous sentences in Indonesian were made by crowdsourcing.
Each Indonesian ambiguous sentence has two possible interpretations, resulting in two disambiguation text outputs for each ambiguous sentence.
Each disambiguation text is made up of two sentences. All of the sentences have been checked by linguists.
