“Deep Speech” From Baidu’s Silicon Valley AI Lab Recognizes Both English and Mandarin With …

Deep Learning System Validates End-to-End Approach to Speech Recognition, Replacing Hand-Engineered Components With Neural Networks SUNNYVALE, CA–(Marketwired – Dec 9, 2015) – Baidu Research, a division of Baidu, Inc. (NASDAQ: BIDU), today unveiled new research results from its Silicon Valley AI Lab (SVAIL). Results include the ability to accurately recognize both English and Mandarin with a single learning algorithm. The results are detailed in a paper posted on ArXiv.org: Deep Speech 2: End-to-End Speech Recognition in English and Mandarin. SVAIL’s Deep Speech system, announced last year, initially focused on improving English speech recognition accuracy in noisy environments (for example, restaurants, cars and public transportation). Over the past year, SVAIL researchers improved Deep Speech’s performance in English and also trained it to transcribe Mandarin. The Mandarin version achieves high accuracy in…


Link to Full Article: “Deep Speech” From Baidu’s Silicon Valley AI Lab Recognizes Both English and Mandarin With …