Mirex 2010: Audio Tag Classification Using Posterior Weighted Bernoulli Mixture Models
نویسندگان
چکیده
Music tags describe different types of semantic information of music. In this paper, we present our submission to the audio tag classification task in MIREX 2010. We propose a posterior weighted Bernoulli mixture model (PWBMM) to automatically annotate a song with tags. The PWBMM approach uses a Gaussian mixture modelbased posterior representation to characterize the 70dimensional music feature vectors of a song and incorporates the counts of tags annotated to the training songs in training the Bernoulli mixture model.
منابع مشابه
Audio Mood Classification Using Ensemble Classifier with Music Tag Based Indexing
This paper presents a system for audio classification using multiple binary ensemble classifiers with music tag based indexing in a one-versus-all classification scenario. The proposed system has won the audio classification task on mood dataset in MIREX 2010 and is implemented as follows. First, in the training phase, the frame-based 70-dimensional feature vectors are extracted from a training...
متن کاملAudio Classification Using Semantic Transformation and Classifier Ensemble
This paper presents our winning audio classification system in MIREX 2010. Our system is implemented as follows. First, in the training phase, the frame-based 70dimensional feature vectors are extracted from a training audio clip by MIRToolbox. Next, the Posterior Weighted Bernoulli Mixture Model (PWBMM) is applied to transform the frame-decomposed feature vectors of the training song into a fi...
متن کاملMirex 2010: Audio Classification Using Semantic Transformation and Classifier Ensemble
Our audio classification system is implemented as follows. First, in the training phase, the frame-based 70 dimensional feature vectors are extracted from a training song by MIRToolbox. Next we apply the Posterior Weighted Bernoulli Mixture Model (PWBMM) to transform the frame-decomposed feature vectors of the training song into a semantic representation based on the music tags; this procedure ...
متن کاملPosterior Weighted Bernoulli Mixture Model for Music Tag Annotation and Retrieval
Music tags provide different types of semantic information about music. Recently, automatic music tagging has generated a great deal of interest among researchers in the field of music information retrieval. In this paper, we propose a posterior weighted Bernoulli mixture model (PWBMM) that automatically annotates a song with tags, or retrieves relevant songs given a semantic tag. The PWBMM app...
متن کاملMirex 2010: Improving Audio Tag Classification by Additional Features and Stacking Svm
This paper describes our system submitted to MIREX 2010 audio tag classification task. The improvements of the current system over our system for MIREX 2009 are as follow. First, additional features are used, such as the event density, pulse clarity, key strength, and some emotion-related features, together with the slope for the frame-decomposed features. Second, the length of each segment is ...
متن کامل