High-level TV talk show structuring centered on speakers’ interventions

Archives professionals have high expectations for efficient indexing tools. In particular, the purpose of archiving TV broadcasts has created an expanding need for automatic content structuring methods. In this chapter, is addressed the task of structuring a particular type of TV content that has been scarcely studied in previous works, namely talk show programs. The object of this work is examined in the light of a number of sociological studies, with the aim to identify relevant prior knowledge on the basis of which the approach followed is motivated. Of particular interest is the fact that, while these broadcasts show a wide disparity of aspects, they also share common features that have been studied in semiological works. For instance, the presence of guests, the polymorphy of the program (a succession of interviews, performances, TV reports, etc.) or the role of the anchorman: using very specific and well-defined codes. These common features, but also the differences, are highlighted by a comparison between two French talk shows from two different periods: “Le Grand Échiquier” (1980s) and “On n’a pas tout dit” (2010s). Building upon that effort and clarifying this rather vague notion of structuring, specific use cases are considered (e.g. the creation of a table of contents for efficient browsing). A taxonomy of the talk show components is then presented, including concepts such as the guests, the hosts, the performances, etc. that suggests a generic high-level structure retaining the main cross-program invariances. These considerations stress out the relevance of considering the speakers’ interventions as elementary units in place of the video shots usually employed