[Accessibility] Re: Sentence Boundary Detection (Chunking)
pdm at brailcom.org
Mon Dec 20 05:03:42 PST 2004
>>>>> "GC" == Gary Cramblitt <garycramblitt at comcast.net> writes:
GC> In a recent discussion on a common speech API, we mentioned that
GC> we ought to develop a standard library for "chunking" text into
GC> sentences, which could be used by speech synthesis authors.
More precisely, we need utterance chunking, not only sentence chunking.
Consider long sentences, which must be split in smaller pieces. Also,
don't forget long sentences may contain long parts without punctuation
(or the input text may not contain any punctuation at all).
It's amazing how much better you feel once you've given up hope.
More information about the Accessibility