This paper introduces a project of advanced system of music retrieval from the Internet. The system uses combination of text search (by author, title and other information about the music file included in id3 tag description or similar for other file types) with more intuitive and novel method of melody search using query by humming. The patterns for storing text and melody information as well as improved clustering algorithm for the pattern space were proposed. The search engine is planned to optimise the query due to the data input by user, thanks to the structure of text and melody index database. The system is planned to be a plug-in for popular digital music players or an independent player. An advanced system of recommendation based on information gathered from user's profile and search history is an integral part of the system. The recommendation mechanism uses scrobbling methods and is responsible for making suggestions of songs unknown to the user but similar to his preferred music styles and positioning search results.