This paper addresses language in its most natural form – in the form of spoken multichannel discourse. It includes the verbal component, prosody, eye gaze, as well as the different kinetic aspects of communication – facial, head, hand and torso gestures. To explore natural multichannel discourse as is, we created a resource “Russian Pear Chats and Stories”. The resource includes 40 sessions with 160 Russian native speakers aged 18–36, 60 men and 100 women; it consists of 15 hours of recording and about 170,000 words. This paper details how the corpus is created and how it can be used. First, we provide an overview of the methodology of multimodality and multichannel corpora. Then we describe the properties of our resource – the data collection set up, the recording software, types of annotation, as well as some avenues of (future) research, including: prosody as an interface between the vocal and gestural channels, specific nature and degree of coordination between manual gestures and elementary discourse units, individual variation and the “portrait” methodology, language production and comprehension in face-to-face communication, and visual attention in natural communication. In its current version, the corpus is available to the scientific community at the project website multidiscourse.ru (in Russian).