Archive for September 27th, 2011

September 27, 2011

Google Tashkeel for diacritics in Arabic to be discontinued

The name Tashkeel – تشكيل means to “give shape or form”. The process of diacritizing is also called Tashkeel.

Google Tashkeel adds missing diacritics to Arabic text:

Diacritic symbols are crucial to identify how words are pronounced and to disambiguate their meanings. Arabic uses diacritic symbols to specify short vowels.

Google Tashkeel Translation product

Google Tashkeel for Arabic diacritic annotation

The symbols are usually omitted by native speakers when writing, as the word meaning can be inferred by context. However, including diacritics is necessary as a pre-processing step for many text processing applications. Diacritics are used in a similar context in many other languages. The two that I am most familiar with are Chinese and Hebrew.

Google is shutting down Tashkeel

Tashkeel will go offline by September 30, 2011. The URL is

http://tashkeel.googlelabs.com

Arabic language by geography

which explains why Tashkeel is shutting down: It is yet another product closing as part of the Google corporate decision to shutter Google Labs.

Tashkeel will be missed by many users. As of September 16, 2011, the product page had over 200 five-star reviews.