题目内容

The following classification statement about speech recognition is wrong ()

A. Voice recognition can be classified according to the type of recognizer, the adaptability of the recognizer to the user, and the size of the voice vocabulary.
B. Continuous word recognition belongs to the classification according to the type of recognizer. With more vocabulary as the object, each word can be completely recognized.
C. According to the adaptation of the recognizer to the user, it can be divided into specific person speech recognition and non-specific person speech recognition. The difference is that the former standard template or model is only suitable for a certain person, and the latter is suitable for a specified one. Speakers of the category (such as standard Mandarin), standard templates or models.
D. The full syllable recognition in the classification according to the size of the voice vocabulary is the basis for achieving unlimited vocabulary or Chinese text input.

查看答案

搜索结果不匹配？点我反馈

更多问题

Natural language includes human communication language, spoken language, written language, sign language, flag language, etc., and artificial language is a natural language improved on the basis of natural language. ()

Support vector machine is one of the most popular and widely discussed machine learning classification methods. This method is suitable for high-dimensional space (there are many features in the feature vector), and can be effectively used for small data sets.

The RL of the direct strategy directly optimizes the objective function and parameterizes the strategy. Compared with the value function, the method of stratified parameters is simpler and easier to converge.

The working principle based on corpus technology is to use statistics to build a bilingual control corpus containing various sentence patterns. ()