Stanislaw Andrzej Raczynski

Polyphonic Pitch Estimation and Instrument Identification by Joint Modeling of Sustained and Attack Sounds

Polyphonic pitch estimation and musical instrument identification are some of the most challenging tasks in the field of music information retrieval (MIR). While existing approaches have focused on the modeling of harmonic partials, we design a joint Gaussian mixture model of the harmonic partials and the inharmonic attack of each note. This model encodes the power of each partial over time as well as the spectral envelope of the attack part. We derive an expectation-maximization (EM) algorithm to estimate the pitch and the parameters of the notes. We then extract timbre features both from the harmonic and the attack part via principal component analysis (PCA) over the estimated model parameters. Musical instrument recognition for each estimated note is finally carried out with a support vector machine (SVM) classifier. Experiments conducted on mixtures of isolated notes as well as real-world polyphonic music show higher accuracy over state-of-the-art approaches based on the modeling of harmonic partials only.

Multipitch estimation by joint modeling of harmonic and transient sounds

Multipitch estimation techniques are widely used for music transcription and acquisition of musical data from digital signals. In this paper, we propose a flexible harmonic temporal timbre model to decompose the spectral energy of the signal in the time-frequency domain into individual pitched notes. Each note is modeled with a 2-dimensional Gaussian mixture. Unlike previous approaches, the proposed model is able to represent not only the harmonic partials but also the inharmonic attack of each note. We derive an Expectation-Maximization (EM) algorithm to estimate the parameters of this model and illustrate the higher performance of the proposed algorithm than NMF algorithm [9] and HTC algorithm [10] for the task of multipitch estimation over synthetic and real-world data.

Note detection with dynamic bayesian networks as a postanalysis step for NMF-based multiple pitch estimation techniques

In this paper we present a method for detecting note events in the note activity matrix obtained with Nonnegative Matrix Factorization, currently the most common method for multipitch analysis. Postprocessing of this matrix is usually neglected by other authors, who use a simple thresholding, often paired with additional heuristics. We propose a theoretically-grounded probabilistic model and obtain very promising results due to the fact that it was able to capture basic musicological information. The biggest advantage of our approach is that it can be extended without much effort to include various information about musical signals, such as principles of tonality and rhythm.

Genre-Based Music Language Modeling with Latent Hierarchical Pitman-Yor Process Allocation

In this work we present a new Bayesian topic model: latent hierarchical Pitman-Yor process allocation (LHPYA), which uses hierarchical Pitman-Yor process priors for both word and topic distributions, and generalizes a few of the existing topic models, including the latent Dirichlet allocation (LDA), the bigram topic model and the hierarchical Pitman-Yor topic model. Using such priors allows for integration of n-grams with a topic model, while smoothing them with the state-of-the-art method. Our model is evaluated by measuring its perplexity on a dataset of musical genre and harmony annotations 3 Genre Database (3GDB) and by measuring its ability to predict musical genre from chord sequences. In terms of perplexity, for a 262-chord dictionary we achieve a value of 2.74, compared to 18.05 for trigrams and 7.73 for a unigram topic model. In terms of genre prediction accuracy with 9 genres, the proposed approach performs about 33% better in relative terms than genre-dependent n-grams, achieving 60.4% of accuracy.

Adaptive prediction order scheme for AMR-WB+

In this paper, we present an adaptive linear prediction order scheme - a simple and effective method to facilitate variable rate coding. We have applied it to the Extended Adaptive Multi-Rate Wideband (AMR-WB+), a state-of-the-art generic speech and audio coder with fixed-length coding designed for wireless transmission. By introducing this scheme into the codec, bit rate can be reduced with a negligible degradation of the quality. The order of linear prediction is adapted based on the audio content, and can be determined by a closed-loop or an open-loop optimal order selection module. The resulted audio quality is examined by average segmental SNR and Perceptual Evaluation of Audio Quality (PEAQ) measure. For this evaluation, speech, music and mixed content audio signals are used. The result proves the effectiveness of this method.