Google search by Voice. For each key area of acoustic modeling and languagemodeling we will describe some of the challenges we faced as well as someof the solutions we have developed to address those unique challenges.In Section 3.1 we will review some of the common metrics we use toevaluate the quality of the recognizer. In Sections 3.2, 3.3, 3.4, we describethe algorithms and technologies used to build the recognizer for Googlesearch by Voice.
đang được dịch, vui lòng đợi..
