Bird Language

ongoing project, 2018-2019 / neural network

The first 15 seconds is a real nightingale singing, the rest part is the singing generated by AI
Bird Language is a project exploring the possibilities of Artificial Intelligence within the context of bio-semiotics. I was interested in the language of birds because it offers in several aspects the nearest analogy to human language and, probably, the imitation of the birds' language was the basis for the human language evolution.
The project is inspired by the ideas of Noam Chomsky about innate linguistic structures - universal grammar - typical for both human language and animals and birds sign systems. The experience with machine learning and text demonstrates that AI captures this universal grammar by statistically extracting the patterns of language, therefore, in the case of the birds' language, machine learning can distinguish peculiar "birds morphemes" - semantic elements, and form from these elements a certain utterance.
The Artificial Intelligence is looking for patterns within birds' sounds to build a mathematical model of the Universal Grammar of Bird Language.
In the first stage of the project, we trained a neural network on the sounds of nightingales to create communication between non-human agents: birds and Artificial Intelligence.
This is a metaphor for communication between nature and technology in which a human being is not necessary.
The second stage of the project is a creation of an AI-translator from bird language to human language.
We started to work with the Great Tit, one of the most widespread species from Europe to Asia.
The first machine learning approach we used, was XGBoost - boosted decision tree algorithms. While it classified bird signs in two groups: calls and songs with an accuracy of 83%, it was not a universal tool to understand the structure of the language.
Now we are developing the second machine learning approach which allows us to see clusters. These clusters reveal the language structure and let us deconstruct the bird language into a series of phonemes, which we can use to build an AI-translator for interspecies communication.
In this case AI is not only a mediator or interface between human beings and birds but rather an organ or full partner, semiotically active. It helps to understand bird's subjectivity through the language.
The inspiration came from Jakob Johann von Uexküll's concept of Umwelt - the experiential world that a being inhabits, its perceptual surrounds.
At this stage we are working in collaboration with scientists: ornithologists and Artificial Intelligence experts.

Credits:

Veronica Samotskaya, ornithologist, scientist-biologist, science popularizer, journalist
Natalia Soboleva, AI expert
Konstantin Yakovlev, PhD in Physics and Mathematics, AI expert
Nikita Prudnikov, musician, developer, AI expert

The project was supported by
Garage Museum of Contemporary Art (Moscow)