Baidu claims deep learning breakthrough with Deep Speech
Baidu says it has developed a speech recognition system, called Deep Speech, the likes of which has never been seen, especially in noisy environments. In restaurant settings and other loud places where other commercial speech recognition systems fail, the deep learning model proved accurate nearly 81 percent of the time.
That might not sound too great, but consider the alternative: commercial speech-recognition APIs against which Deep Speech was tested, including those for Microsoft Bing, Google and Wit.AI, topped out at nearly 65 percent accuracy in noisy environments.
lucy24
6:22 am on Dec 24, 2014 (gmt 0)
Why does an East Asian search engine want to become adept at recognizing spoken English (the only language used in the article's examples)? I want to see research comparing computers' speech recognition among different human languages.
bill
10:27 am on Dec 24, 2014 (gmt 0)
Baidu Research [research.baidu.com...] , which apparently is working on this, is based in Silicon Valley and Beijing. I wouldn't be too surprised if they started work in English as they have a lot of Stanford and other US university ties. Perhaps speech recognition technology is more advanced in English? Not sure. But Baidu is entering a lot of non-Chinese markets as well, so it does fit the pattern.
