Editing Bigram (section)

==Applications==

Bigrams, along with other n-grams, are used in most successful [[language model]]s for [[speech recognition]].<ref>{{cite book |last1=Collins |first1=Michael John |title=Proceedings of the 34th annual meeting on Association for Computational Linguistics - |chapter=A new statistical parser based on bigram lexical dependencies |date=1996-06-24 |pages=184–191 |doi=10.3115/981863.981888 |chapter-url=http://www.aclweb.org/anthology/P96-1025 |access-date=2018-10-09 |publisher=Association for Computational Linguistics|arxiv=cmp-lg/9605012 |s2cid=12615602 }}</ref>

Bigram frequency attacks can be used in [[cryptography]] to solve [[cryptograms]]. See [[Frequency analysis (cryptanalysis)|frequency analysis]].

Bigram frequency is one approach to [[Language detection#Statistical approaches|statistical language identification]].

Some activities in [[logology (linguistics)|logology]] or recreational linguistics involve bigrams. These include attempts to find English words beginning with every possible bigram,<ref>{{cite journal|last=Cohen|first=Philip M.|year=1975|title=Initial Bigrams|journal=Word Ways|volume=8|issue=2|url=http://digitalcommons.butler.edu/wordways/vol8/iss2/8 |access-date=11 September 2016}}</ref> or words containing a string of repeated bigrams, such as ''logogogue''.<ref>{{cite journal|last=Corbin|first=Kyle|year=1989|title=Double, Triple, and Quadruple Bigrams|journal=Word Ways|volume=22|issue=3|url=http://digitalcommons.butler.edu/wordways/vol22/iss3/8 |access-date=11 September 2016}}</ref>