Presentation of spoken data content is based on the annotation system designed to fulfil the attempted research goals. It should include the orthographic transcription, the linguistic annotation such as markings of extragrammatical spontaneous speech sequences, and the documentation of data collection and processing. Especially, spontaneous conversation contains a great variety of sentential construction, pronunciation variation, and conversational interaction. This paper gives an overview of the linguistic annotation system and the tool we have developed for transcribing Mandarin spontaneous conversational dialogues. By means of the annotated data, a character-based linguistic database is construed to provide quantified materials for queries.