nltk.PunktSentenceTokenizer.span_tokenize¶

PunktSentenceTokenizer.span_tokenize(text, realign_boundaries=True)[source]¶

Given a text, returns a list of the (start, end) spans of sentences in the text.