On the Construction of Knowledge-Communication-Oriented Computer Spoken Corpus for Endangered Languages

This Corpus/corpora is a computer database that stores language materials. Corpus in the world is dominated by lingua franca corpus. However, these corpora have limited samples and are difficult to meet the needs of machine artificial intelligence. The paper proposes to build a Knowledge-Communication-Oriented spoken corpus for Endangered Languages to enrich the content of corpus research. A Knowledge-Communication-Oriented Spoken Corpus was divided into three sub-corporas, the words sub-corpu, the sentences sub-corpus, and the narrative discourses sub-corpus. This paper mainly introduces the methods of constructing the spoken corpus of endangered languages from the aspects of corpus collection, corpus arrangement and corpus annotation.