A team of Finnish researchers has developed a new prototype of "smart" radio for smartphones that will provide customised songs to each listener's preferences.
The recommendation engine developed by the VTT Technical Research Centre in Finland recognises the user's preferences from small actions they perform while listening such as liking a song or interrupting one.
It uses this information to recommend suitable content from the radio's archives and other sources.
"We aim to create a user-friendly radio for mobile phones that provides content tailored to each listener's preferences," VTT Technical Research Centre said in a statement on Friday.
The recommendation engine service utilises speech recognition and keywords to identify similarities between talk shows.
If, for example, certain keywords were mentioned in a programme prior to an action made by the listener, the recommendation engine concludes that the action is related to these words and that the listener likes or dislikes the programme.
The service does not jeopardise the listener's privacy as it does not use the listeners' personal information or information about their interests or hobbies.
The service has been developed for the Android mobile operating system but it can be adapted to other platforms too.
For radio stations, the "smart" radio opens up the opportunity of providing more versatile and targeted programmes, the statement added.