Swedish scientists hope to establish communication with bottlenose dolphins. To do this, the dolphin signals will be analyzed by an algorithm that works with 40 natural languages.
Dolphins, like other representatives of the cetacean order, have a developed communication system. To communicate with each other, animals use sign language (body positions, jumps, turns) and a variety of sounds. The most common signals of dolphins are whistles, scientists distinguish about 200 different types of such sounds. To record dolphin signals, hydrophones are used - special microphones that can record sound and ultrasound underwater.
In 2016, researchers found that bottlenose dolphins have individual "names." These are specific sequences of whistles, clicks, and other signals by which animals call each other. Scientists recorded these sequences and later played back the recording - the dolphins only responded to their "name".
Scientists have repeatedly tried to detect characteristic patterns in the signals of dolphins and thus decipher the "speech" of animals. The first experiments took place in the middle of the 20th century: ethologists Evans and Bastian found that dolphins can transmit meaningful information to each other. The experiments of Evans and Bastian were based not on the analysis of natural vocalizations of animals, but on the work with special equipment. Dolphins, located in two closed areas of the pool, gave each other a signal by pressing on the pedals. For the correct answer, animals received a reward. Later, this experience was repeated by other groups of researchers.
Modern research into dolphin communication is often based on the analysis of a large array of recorded sounds. Scientists from the Royal Institute of Technology in Stockholm plan to carry out such work over the next four years. The material for the study will be the sounds of bottlenose dolphins living in the Kolmorden Dolphinarium in southern Sweden. A self-learning algorithm developed by Gavagai AB will analyze the signals. This algorithm has been used to create mathematical models describing about 40 natural languages. One of the tasks of such algorithms is to create electronic dictionaries that can be used for automatic translation or spell checking. Scientists plan to create such a "dictionary" for dolphins: the algorithm must learn to recognize the characteristic sequences of signals emitted by bottlenose dolphins,and build a connection between them.
In 2016, researchers at the Karadag Nature Reserve recorded and analyzed the signals of two bottlenose dolphins. Specially developed technology made it possible to distinguish the "voices" of dolphins. Scientists have suggested that bottlenose dolphins are able to form "words" from individual phonemes and combine them into "sentences" of up to five "words." According to the researchers, the creation of automated systems for analyzing the language of dolphins will help to learn much more about the animal communication system.
Dolphin sounds recorded in the Karadag reserve. I - signals from male Yasha, II - signals from female Yana. St. Petersburg Polytechnical University Journal: Physics and Mathematics
Natalia Pelezneva
Promotional video: