As a technology company, one of the serious work with Google is investing in the field of localization of language making it easier to use users various countries. In Indonesia, Google promised to continue to provide services better in recognizing the Indonesian language.
Included in the services the new Google Voice Search is officially introduced in Jakarta, Wednesday (03/16/2011). This is a voice-based search layananj where the User is to say what is sought and will appear in Internet search results. This machine is not guaranteed to always be right 100 percent recognize the Indonesian language, but can learn.
"Of course depending on how the user saying the search terms. If the pronunciation is clear in the direct understanding and out of search results accordingly," Henk said Prihatna, Indonesia Country Consultant, Google in Jakarta, Wednesday (16/03/2011). If you do not find a suitable sentence or phrase, the machine will provide a list of phrases that are similar recommendation. Once chosen, the Google engine will automatically learn new phrases. That's how Google learning Indonesian.
However, before Google can learn, before it launched the service had already learned to recognize words, sentence structure and grammar Indonesian. In fact, exceptions are commonly found in Bahasa Indonesian.
"Many challenges to localize services in Indonesian," he said. First, Indonesian is homograf writing the same word but different pronunciation and meaning as the word apple (ceremony) and apple (fruit), voice search should be able to distinguish
Second, Indonesian and even after there EYD (ajaan enhanced) still affected many Dutch words. For example the name of Sukarno Sukarno and the same pronunciation but different writings. Voice search should be able to overcome it.
Third, some people use the phrase mixture Indonesia Indonesian and English. Services must be able to menegenali words in different languages in one sentence.
Fourth, there is currently no standardization of the pronunciation of words in Bahasa Indonesian. EYD new to writing, but to the pronunciation depends on access so that prononciation vary depending on ethnic origin or region of origin of speakers.
"Google since last year to recruit and provide time for the engineer to the data collected on," said Heng. He said one way is to involve as many as 500 student volunteers. They were asked to pronounce the word list so that the collected database to hundreds of thousands of phrases. This database is then stored in a server cloud that will translate the search text.
According to the story Heng directly involved in the project, he also sent to one of Google's research center in New York to improve the model's voice search in Indonesian language. Machine duji try to do a search with voice over and over again until the level of truth as possible. And over time, machine learning new phrases from many users of the service.
source: kompas.com
0 comments:
Post a Comment