Hummer Specific Learning to Enhance Query-by-Humming
نویسندگان
چکیده
The goal of this project is to evaluate how well hummer-specific information can enhance the query-by-humming scheme developed by Bryan Pardo and David Little. We use a genetic algorithm to optimize six parameters to the query-by-humming system over queries by single individuals as compared to optimizing those parameters over queries by all the hummers. We performed an experiment with 24 queries of 6 songs sung by 4 singers, with a set of 1001 targets. Our experiments showed that our QBH system performs better than chance, and that learned parameters are better than default human-selected parameters, but in our experiment, learning over multiple individuals performed better than learned for specific singers.
منابع مشابه
Prototyping a Vibrato-Aware Query-By-Humming (QBH) Music Information Retrieval System for Mobile Communication Devices: Case of Chromatic Harmonica
Background and Aim: The current research aims at prototyping query-by-humming music information retrieval systems for smart phones. Methods: This multi-method research follows simulation technique from mixed models of the operations research methodology, and the documentary research method, simultaneously. Two chromatic harmonica albums comprised the research population. To achieve the purpose ...
متن کاملA Machine Learning based Music Retrieval and Recommendation System
In this paper, we present a music retrieval and recommendation system using machine learning techniques. We propose a query by humming system for music retrieval that uses deep neural networks for note transcription and a note-based retrieval system for retrieving the correct song from the database. We evaluate our query by humming system using the standard MIREX QBSH dataset. We also propose a...
متن کاملA Note Based Query By Humming System Using Convolutional Neural Network
In this paper, we propose a note-based query by humming (QBH) system with Hidden Markov Model (HMM) and Convolutional Neural Network (CNN) since note-based systems are much more efficient than the traditional frame-based systems. A note-based QBH system has two main components: humming transcription and candidate melody retrieval. For humming transcription, we are the first to use a hybrid mode...
متن کاملA Music Retrieval System with Spelling Correction Technique
In this paper, we describe a novel music information retrieval system using spelling correction technique. The current music retrieval systems with humming enable a user give a query with humming. However, there is a problem that a user cannot always input a complete melody to the system with humming. It is because the user often mistakes a melody in its query. In order to solve the problem, we...
متن کاملQuery by Singing/Humming System Based on Deep Learning
With the proliferation of digital music, efficient indexing and retrieval tools are required for searching the desired music in a large digital music database (DB). Traditional text-based information retrieval methods (titles, lyrics, singers, etc.) cannot meet people’s needs now. Music information retrieval (MIR) has been a matter of interest. Therefore, how to retrieve the music information q...
متن کامل