Automatic identification of dialects

Goal

• In a continuous acoustic stream of mixed dialects, to be able to identify dialect sections
• To do the above primarily acoustically

Experimental setup

A mix of the following dialects is being created from examples of arabic from the following table. This will be the basis of dialect identification experiments.

Afghanistan	algerian	algerian saharan spoken	algerian saharan: tamanrasset	algerian spoken
algerian:constantine	baghdadi	cairene	chad	chadian
chadian spoken	chadian: shuwa	chadic	classical	damascus
eastern algerian	egyptian	egyptian : cairene	egyptian colloquial	guera group
hassaniya	high	n. levantine spoken	nomad	normal egyptian
north lev antine: beiruti	north levantine: central syrian damascas	north levantine: north lebanese literary	saharan	south levantine spoken
north levantine: syrian calamon	north syrian	northern coll. sudanese	northern yemeni	palestanian-jordanian
palestinian	palestinian-jordanian	pidgin	quranic	sa`idi
sanaani	shua	shuwa	shuwa: guera	south levantine
south levantine: jordanian	south levantine: palestinian	southern coll. sudanese	southern levantine	tamanghasset
tamanrasset	tunisian	yemen	iraqi	iraqi judeo-
jewish iraqi-baghdadi	juba	judeo-iraqi	khartoum	koranic
lebanese-syrian	lebanon	levantine	lower egypt	maghrebi
maghribi colloquial	mesopotamian gelet	mesopotamian qeltu	mesopotamian spoken	mesopotamian: baghdadi
middle east	modern literary	modern standard	moroccan	moroccan : moroccan colloquial
moroccan colloquial	moroccan: jebli moroccan spoken	morocco	north levantine	north levantine spoken
southern sudan	standard	sudanese creole	sudanese creole: juba	sudanese spoken
sudanese: north kordofan	syrian : dayr 'atiyah	syro-lebanese	ta'izzi-adeni: adeni	ta'izzi-adeni: el odei
ta'izzi-adeni: hojeria	ta'izzi-adeni: jibla	ta'izzi-adeni: tihami	ta'izzi-adeni: zabid	ta`izzi-adeni
ta`izzi-adeni: adeni	ta`izzi-adeni: el odein	ta`izzi-adeni: hojeria	ta`izzi-adeni: jibla	ta`izzi-adeni: ta`izzi
ta`izzi-adeni: tihami	ta`izzi-adeni: zabid	tajiki spoken	tajiki: afghanistan	yemen: sana`a
yemeni	sudanese: khartoum

-- RitaSingh - 08 Apr 2008