The Chinese company Baidu, which owns the eponymous search engine, has been developing its own artificial intelligence for several years. Research began back in 2013, and now the AI from Baidu shows very good results. For example, he recently learned to speak without the help of a person. Recently, the company introduced a text-to-speech system called Deep Voice.
The neural network used the principles of deep learning to analyze speech and compare it with text, after which it began quite tolerably to pronounce words. First, Deep Voice decomposes the word into separate sounds, then enters the data into the speech synthesis system and utters the desired word. The results are far from perfect, so Baidu has a lot of work to do.
Despite the significant success of the Chinese search giant, Google continues to lead in this area. Her system WaveNet already successfully synthesizes speech, based on the previously “read” text. Now the American development is able to speak two languages, and experts continue to work on it. At the moment, Google is hiring new employees, planning to increase the staff of developers twice.
The article is based on materials .
Comments
Post a Comment