The paper presents the description of semiotic approach to studying daily life spoken language. The research is being carried out on the material of the ORD corpus, which is developed in St. Petersburg State University. The corpus is being created with the aim to study Russian spontaneous speech in natural communicative situations. Its collections contain more than 1250 hours of speech recordings, made by volunteers-participants, and represent valuable research data for many interdisciplinary studies like anthropological linguistics, behavioral studies, studies in pragmatics, discourse analysis, psycholinguistics, etc. The methodology of the given research is based on two approaches. The first of them is based on speculative understandings of the main concepts of daily-life spoken communication. Following this approach, the recorded conversations are to be annotated in accordance with this scheme, having as a result the statistics of their distribution. The second approach is "a technological one" and is based on the usage of specialized computer software. These programs allow to build conceptual nets peculiar to different communicative situations and to various types of participants.