随着相关学科及计算机技术的发展，多学科、多维度、多元化、多功能的语音数据库具有广阔的发展前景。本文拟以汉语方言异读词为研究对象，主要是文白异读和新老异读，运用社会语言学、历史比较语言学、计算语言学、地理语言学、语言类型学等理论和方法，设计一个语音数据库，穷尽式地考查汉语方言文白异读和新老异读现象，以期通过大数据的研究方法厘清这两种现象的特点、来源、类型、变迁及二者的区别等问题。
With the development of related subjects and computer technology, multi-disciplinary, mul-ti-dimensional, multi-functional voice database has a broad development prospect. This thesis intends to use Chinese dialect as the research object, mainly for literary and colloquial readings and the old and new variant pronunciation, using sociolinguistics, historical comparative linguistics, computational linguistics, geographical linguistics, linguistic typology and other theories, to design a speech database and exhaustively examine the phenomenon of literary and colloquial readings, and the old and new reading, in order to clarify the characteristics, sources, types, changes and differences between the two phenomena through the research methods of big data.
