Phonological Word Segmentation Task

An Phonological Word Segmentation Task is a Speech Segmentation Task that is also Word Segmentation Task for the Detection of Phonological Words in a Spoken Expression.

• AKA: PWST, Phonological Word Segmentation.
• Context:
  • Input: a Spoken Expression.
  • output: a Phonological Word Sequence.
  • It can require a Spoken Expression that is spoken extremely slowly.
  • It can require Context to Disambiguate Homophones.
• Example(s):
  • PWST(I'mcominghome) ⇒ ([I'm] [coming] [home]), an example of PWST.
  • PWST(haʊtuɹɛkonaɪzbiːtʃ) ⇒ ([haʊ] [tu] [ɹɛk] [o] [naɪz] [biːtʃ]), i.e. “[how/haʊ][to/tu][wreck/ɹɛk][a/a][nice/naɪz][beach]".
  • PWST(haʊtuɹɛkonaɪzbiːtʃ) ⇒ ([haʊ] [tu] [ɹɛkonaɪz] [biːtʃ]), i.e. “[how/haʊ][to/tu][recognize/ɹɛkəgnaɪz][speech/sbiːtʃ]".
  • …
• Counter-Example(s):
  • PWST("I'm coming home") ⇒ ([I'm] [coming] [home]), an example of OWST.
  • PWST("I bought a real time operating system") ⇒ ([I] [bought] [a] [real] [time] [operating] [system]), an example of OWST.
  • PWST("I'm coming home") ⇒ ([I] ['m] [coming] [home]), an example of SWST.
  • PWST("I bought a real time operating system") ⇒ ([I] [bought] [a] [real time] [operating system]), an example of VWST.
  • PWST("日文章魚怎麼說") ⇒ ([日文] [章魚] [怎麼] [說]) (i.e. ~[Japanese] [octopus] [how] [say]), an example of VWST.
• See: Orthographic Word Segmentation Task, IPA.

---

---

References

• (Wikipedia, 2009) ⇒ http://en.wikipedia.org/wiki/Speech_segmentation#Lexical_segmentation

---