Spoken language evaluation method through freely read topics and spoken language evaluation system thereof

Abstract

The invention discloses a spoken language evaluation method through freely read topics and a spoken language evaluation system thereof. The method comprises the steps that voice signals to be evaluated are received; voice recognition is performed on the voice signals, and voice segments corresponding to all the basic voice units of the voice signals are obtained based on segmentation of the limited boundary of a recognition text; the pronunciation accuracy characteristics of the voice signals are extracted according to the recognition text and all the voice segments, and the pronunciation accuracy characteristics act as the evaluation characteristics of the voice signals; the pronunciation accuracy characteristics include any one or multiple of the following: the posterior probability of non-mistakenly recognized words, the proportion of misread words and the proportion of correctly read words; and the evaluation score of the voice signals is calculated according to the evaluation characteristics of the voice signals and an evaluation model which is constructed in advance. Automatic evaluation through the freely read topics can be accurately realized.

Description

technical field [0001] The invention relates to the technical field of speech signals, in particular to a spoken language evaluation method and system for freely reading questions. Background technique [0002] As an important medium of interpersonal communication, spoken language plays an extremely important role in real life. With the continuous development of social economy and the intensification of globalization, people have put forward higher and higher requirements for the efficiency of language learning and the objectivity, fairness and scale of test of language assessment. To this end, the industry has developed some language teaching and evaluation systems. [0003] With the continuous maturity of oral evaluation technology, more and more oral language learning or oral language teachers are using this technology to teach and learn oral English. The current common oral language learning scenario is to specify the text to be read aloud, and then evaluate the accura...

AI Technical Summary

Problems solved by technology

[0004] Because the existing oral English evaluation technology is based on the prior text content of the specified reading text, and there is no standard answer for the free reading question type, the existing oral English evaluation technology cannot accurately realize the automatic evaluation of the free reading question type

Images