In the present project, we propose a pioneer research that will provide new insights into human language ecology and evolution by exploring three rarely studied natural traditional oral practices with a multidisciplinary methodology. These are alternative speech forms that always rely on a spoken language and that consist of mapping voiced speech signals into very different types of sounds, such as whistling, instrumental music and animal calls. These speech transformations are respectively called ‘whistled languages’, ‘the singing/speaking mode of musical instruments’ (or ‘instrumental speech’), and ‘onomatopoetic words named after animal calls’. One of the main focuses of the project will be to analyze the interrelations between production and perception in such emulation processes based on iconicity with voiced speech (with methods of linguistic description as well as behavioral and electrophysiological experiments). Indeed, each of these phenomena transforms speech in a way that highlights some language-specific phonetic and phonological processes as well as some key perceptual processes of human speech recognition. The second principal axe of the project will be to analyze the impact of the biotope on human language because such oral practices are the result of both productive and perceptive adaptations of speech to real ecological situations (both social and biological). We will focus on the impact of the acoustic constraints imposed by typical natural biotopes on such special speech forms (such as dense forests, plain open fields and deserts). Our project will include a study on how such environmental constraints also affect shouted and normal speech, as well as a comparison with animal communication adaptation to the same constraints. The applicant is one of the main specialists on these topics for which he has developed pioneer methodologies. This project will reinforce the Franco-Brazilian collaborations of the international network he has already built.