Modern requirements for comfort and safety in residential, industrial, and transportation spaces are constantly increasing, which stimulates the search for new, more effective, and environmentally friendly acoustic materials. Traditional sound-absorbing and sound-insulating materials are often made from synthetic polymers, such as mineral wool, fiberglass, and expanded polystyrene, which are non-biodegradable and can pose a risk to human health and the environment. Consequently, environmentally friendly sound-absorbing materials based on plant-derived raw materials and biodegradable polymers are gaining increasing popularity. Traditional methods for designing such materials are often based on an empirical approach and require a large number of expensive experiments. However, with the development of machine learning (ML), new opportunities are emerging to optimize this process, allowing for the creation of materials with specified acoustic characteristics more quickly and efficiently. The work analyzes current trends in the selection of component base, technologies, and research approaches in the creation of new acoustic materials.