arXiv subjects

Gerhard Widmer

Publications and source records attributed to Gerhard Widmer.

At least 163 records · Page 9Linked to original sources

arXiv2018

Imposing higher-level Structure in Polyphonic Music Generation using Convolutional Restricted Boltzmann Machines and Constraints

We introduce a method for imposing higher-level structure on generated, polyphonic music. A Convolutional Restricted Boltzmann Machine (C-RBM) as a generative model is combined with gradient descent constraint optimisation to provide further control over the generation process. Among other things, this allows for the use of a "template" piece, from which some structural properties can be extracted, and transferred as constraints to the newly generated material. The sampling process is guided with Simulated Annealing to avoid local optima, and to find solutions that both satisfy the constraints, and are relatively stable with respect to the C-RBM. Results show that with this approach it is possible to control the higher-level self-similarity structure, the meter, and the tonal properties of the resulting musical piece, while preserving its local musical coherence.

cs.SD↗

arXiv2018

A Large-Scale Study of Language Models for Chord Prediction

We conduct a large-scale study of language models for chord prediction. Specifically, we compare N-gram models to various flavours of recurrent neural networks on a comprehensive dataset comprising all publicly available datasets of annotated chords known to us. This large amount of data allows us to systematically explore hyper-parameter settings for the recurrent neural networks---a crucial step in achieving good results with this model class. Our results show not only a quantitative difference between the models, but also a qualitative one: in contrast to static N-gram models, certain RNN configurations adapt to the songs at test time. This finding constitutes a further step towards the development of chord recognition systems that are more aware of local musical context than what was previously possible.

cs.LG↗

arXiv2017

The ACCompanion v0.1: An Expressive Accompaniment System

In this paper we present a preliminary version of the ACCompanion, an expressive accompaniment system for MIDI input. The system uses a probabilistic monophonic score follower to track the position of the soloist in the score, and a linear Gaussian model to compute tempo updates. The expressiveness of the system is powered by the Basis-Mixer, a state-of-the-art computational model of expressive music performance. The system allows for expressive dynamics, timing and articulation.

cs.SD↗

arXiv2017

What were you expecting? Using Expectancy Features to Predict Expressive Performances of Classical Piano Music

In this paper we present preliminary work examining the relationship between the formation of expectations and the realization of musical performances, paying particular attention to expressive tempo and dynamics. To compute features that reflect what a listener is expecting to hear, we employ a computational model of auditory expectation called the Information Dynamics of Music model (IDyOM). We then explore how well these expectancy features -- when combined with score descriptors using the Basis-Function modeling approach -- can predict expressive tempo and dynamics in a dataset of Mozart piano sonata performances. Our results suggest that using expectancy features significantly improves the predictions for tempo.

cs.SD↗

arXiv2017

Music Playlist Continuation by Learning from Hand-Curated Examples and Song Features: Alleviating the Cold-Start Problem for Rare and Out-of-Set Songs

Automated music playlist generation is a specific form of music recommendation. Generally stated, the user receives a set of song suggestions defining a coherent listening session. We hypothesize that the best way to convey such playlist coherence to new recommendations is by learning it from actual curated examples, in contrast to imposing ad hoc constraints. Collaborative filtering methods can be used to capture underlying patterns in hand-curated playlists. However, the scarcity of thoroughly curated playlists and the bias towards popular songs result in the vast majority of songs occurring in very few playlists and thus being poorly recommended. To overcome this issue, we propose an alternative model based on a song-to-playlist classifier, which learns the underlying structure from actual playlists while leveraging song features derived from audio, social tags and independent listening logs. Experiments on two datasets of hand-curated playlists show competitive performance compared to collaborative filtering when sufficient training data is available and more robust performance when recommending rare and out-of-set songs. For example, both approaches achieve a recall@100 of roughly 35% for songs occurring in 5 or more training playists, whereas the proposed model achieves a recall@100 of roughly 15% for songs occurring in 4 or less training playlists, compared to the 3% achieved by collaborative filtering.

cs.IR↗

arXiv2017

Learning Musical Relations using Gated Autoencoders

Music is usually highly structured and it is still an open question how to design models which can successfully learn to recognize and represent musical structure. A fundamental problem is that structurally related patterns can have very distinct appearances, because the structural relationships are often based on transformations of musical material, like chromatic or diatonic transposition, inversion, retrograde, or rhythm change. In this preliminary work, we study the potential of two unsupervised learning techniques - Restricted Boltzmann Machines (RBMs) and Gated Autoencoders (GAEs) - to capture pre-defined transformations from constructed data pairs. We evaluate the models by using the learned representations as inputs in a discriminative task where for a given type of transformation (e.g. diatonic transposition), the specific relation between two musical patterns must be recognized (e.g. an upward transposition of diatonic steps). Furthermore, we measure the reconstruction error of models when reconstructing musical transformed patterns. Lastly, we test the models in an analogy-making task. We find that it is difficult to learn musical transformations with the RBM and that the GAE is much more adequate for this task, since it is able to learn representations of specific transformations that are largely content-invariant. We believe these results show that models such as GAEs may provide the basis for more encompassing music analysis systems, by endowing them with a better understanding of the structures underlying music.

cs.SD↗

arXiv2017

Probabilistic Generative Adversarial Networks

We introduce the Probabilistic Generative Adversarial Network (PGAN), a new GAN variant based on a new kind of objective function. The central idea is to integrate a probabilistic model (a Gaussian Mixture Model, in our case) into the GAN framework which supports a new kind of loss function (based on likelihood rather than classification loss), and at the same time gives a meaningful measure of the quality of the outputs generated by the network. Experiments with MNIST show that the model learns to generate realistic images, and at the same time computes likelihoods that are correlated with the quality of the generated images. We show that PGAN is better able to cope with instability problems that are usually observed in the GAN training procedure. We investigate this from three aspects: the probability landscape of the discriminator, gradients of the generator, and the perfect discriminator problem.

cs.LG↗

arXiv2017

Piece Identification in Classical Piano Music Without Reference Scores

In this paper we describe an approach to identify the name of a piece of piano music, based on a short audio excerpt of a performance. Given only a description of the pieces in text format (i.e. no score information is provided), a reference database is automatically compiled by acquiring a number of audio representations (performances of the pieces) from internet sources. These are transcribed, preprocessed, and used to build a reference database via a robust symbolic fingerprinting algorithm, which in turn is used to identify new, incoming queries. The main challenge is the amount of noise that is introduced into the identification process by the music transcription algorithm and the automatic (but possibly suboptimal) choice of performances to represent a piece in the reference database. In a number of experiments we show how to improve the identification performance by increasing redundancy in the reference database and by using a preprocessing step to rate the reference performances regarding their suitability as a representation of the pieces in question. As the results show this approach leads to a robust system that is able to identify piano music with high accuracy -- without any need for data annotation or manual data preparation.

cs.IR↗

arXiv2017

Learning Audio - Sheet Music Correspondences for Score Identification and Offline Alignment

This work addresses the problem of matching short excerpts of audio with their respective counterparts in sheet music images. We show how to employ neural network-based cross-modality embedding spaces for solving the following two sheet music-related tasks: retrieving the correct piece of sheet music from a database when given a music audio as a search query; and aligning an audio recording of a piece with the corresponding images of sheet music. We demonstrate the feasibility of this in experiments on classical piano music by five different composers (Bach, Haydn, Mozart, Beethoven and Chopin), and additionally provide a discussion on why we expect multi-modal neural networks to be a fruitful paradigm for dealing with sheet music and audio at the same time.

cs.IR↗

arXiv2017

Likelihood Estimation for Generative Adversarial Networks

We present a simple method for assessing the quality of generated images in Generative Adversarial Networks (GANs). The method can be applied in any kind of GAN without interfering with the learning procedure or affecting the learning objective. The central idea is to define a likelihood function that correlates with the quality of the generated images. In particular, we derive a Gaussian likelihood function from the distribution of the embeddings (hidden activations) of the real images in the discriminator, and based on this, define two simple measures of how likely it is that the embeddings of generated images are from the distribution of the embeddings of the real images. This yields a simple measure of fitness for generated images, for all varieties of GANs. Empirical results on CIFAR-10 demonstrate a strong correlation between the proposed measures and the perceived quality of the generated images.

cs.LG↗

arXiv2017

Modeling Harmony with Skip-Grams

String-based (or viewpoint) models of tonal harmony often struggle with data sparsity in pattern discovery and prediction tasks, particularly when modeling composite events like triads and seventh chords, since the number of distinct n-note combinations in polyphonic textures is potentially enormous. To address this problem, this study examines the efficacy of skip-grams in music research, an alternative viewpoint method developed in corpus linguistics and natural language processing that includes sub-sequences of n events (or n-grams) in a frequency distribution if their constituent members occur within a certain number of skips. Using a corpus consisting of four datasets of Western classical music in symbolic form, we found that including skip-grams reduces data sparsity in n-gram distributions by (1) minimizing the proportion of n-grams with negligible counts, and (2) increasing the coverage of contiguous n-grams in a test corpus. What is more, skip-grams significantly outperformed contiguous n-grams in discovering conventional closing progressions (called cadences).

cs.IR↗

arXiv2017

Automatic estimation of harmonic tension by distributed representation of chords

The buildup and release of a sense of tension is one of the most essential aspects of the process of listening to music. A veridical computational model of perceived musical tension would be an important ingredient for many music informatics applications. The present paper presents a new approach to modelling harmonic tension based on a distributed representation of chords. The starting hypothesis is that harmonic tension as perceived by human listeners is related, among other things, to the expectedness of harmonic units (chords) in their local harmonic context. We train a word2vec-type neural network to learn a vector space that captures contextual similarity and expectedness, and define a quantitative measure of harmonic tension on top of this. To assess the veridicality of the model, we compare its outputs on a number of well-defined chord classes and cadential contexts to results from pertinent empirical studies in music psychology. Statistical analysis shows that the model's predictions conform very well with empirical evidence obtained from human listeners.

cs.SD↗

arXiv2017

A Hybrid Approach with Multi-channel I-Vectors and Convolutional Neural Networks for Acoustic Scene Classification

In Acoustic Scene Classification (ASC) two major approaches have been followed . While one utilizes engineered features such as mel-frequency-cepstral-coefficients (MFCCs), the other uses learned features that are the outcome of an optimization algorithm. I-vectors are the result of a modeling technique that usually takes engineered features as input. It has been shown that standard MFCCs extracted from monaural audio signals lead to i-vectors that exhibit poor performance, especially on indoor acoustic scenes. At the same time, Convolutional Neural Networks (CNNs) are well known for their ability to learn features by optimizing their filters. They have been applied on ASC and have shown promising results. In this paper, we first propose a novel multi-channel i-vector extraction and scoring scheme for ASC, improving their performance on indoor and outdoor scenes. Second, we propose a CNN architecture that achieves promising ASC results. Further, we show that i-vectors and CNNs capture complementary information from acoustic scenes. Finally, we propose a hybrid system for ASC using multi-channel i-vectors and CNNs by utilizing a score fusion technique. Using our method, we participated in the ASC task of the DCASE-2016 challenge. Our hybrid approach achieved 1 st rank among 49 submissions, substantially improving the previous state of the art.

cs.SD↗

arXiv2017

End-to-End Musical Key Estimation Using a Convolutional Neural Network

We present an end-to-end system for musical key estimation, based on a convolutional neural network. The proposed system not only out-performs existing key estimation methods proposed in the academic literature; it is also capable of learning a unified model for diverse musical genres that performs comparably to existing systems specialised for specific genres. Our experiments confirm that different genres do differ in their interpretation of tonality, and thus a system tuned e.g. for pop music performs subpar on pieces of electronic music. They also reveal that such cross-genre setups evoke specific types of error (predicting the relative or parallel minor). However, using the data-driven approach proposed in this paper, we can train models that deal with multiple musical styles adequately, and without major losses in accuracy.

cs.LG↗

arXiv2017

On the Futility of Learning Complex Frame-Level Language Models for Chord Recognition

Chord recognition systems use temporal models to post-process frame-wise chord preditions from acoustic models. Traditionally, first-order models such as Hidden Markov Models were used for this task, with recent works suggesting to apply Recurrent Neural Networks instead. Due to their ability to learn longer-term dependencies, these models are supposed to learn and to apply musical knowledge, instead of just smoothing the output of the acoustic model. In this paper, we argue that learning complex temporal models at the level of audio frames is futile on principle, and that non-Markovian models do not perform better than their first-order counterparts. We support our argument through three experiments on the McGill Billboard dataset. The first two show 1) that when learning complex temporal models at the frame level, improvements in chord sequence modelling are marginal; and 2) that these improvements do not translate when applied within a full chord recognition system. The third, still rather preliminary experiment gives first indications that the use of complex sequential models for chord prediction at higher temporal levels might be more promising.

cs.SD↗

arXiv2017

An Experimental Analysis of the Entanglement Problem in Neural-Network-based Music Transcription Systems

Several recent polyphonic music transcription systems have utilized deep neural networks to achieve state of the art results on various benchmark datasets, pushing the envelope on framewise and note-level performance measures. Unfortunately we can observe a sort of glass ceiling effect. To investigate this effect, we provide a detailed analysis of the particular kinds of errors that state of the art deep neural transcription systems make, when trained and tested on a piano transcription task. We are ultimately forced to draw a rather disheartening conclusion: the networks seem to learn combinations of notes, and have a hard time generalizing to unseen combinations of notes. Furthermore, we speculate on various means to alleviate this situation.

cs.SD↗

arXiv2016

Towards Score Following in Sheet Music Images

This paper addresses the matching of short music audio snippets to the corresponding pixel location in images of sheet music. A system is presented that simultaneously learns to read notes, listens to music and matches the currently played music to its corresponding notes in the sheet. It consists of an end-to-end multi-modal convolutional neural network that takes as input images of sheet music and spectrograms of the respective audio snippets. It learns to predict, for a given unseen audio snippet (covering approximately one bar of music), the corresponding position in the respective score line. Our results suggest that with the use of (deep) neural networks -- which have proven to be powerful image processing models -- working with sheet music becomes feasible and a promising future research direction.

cs.LG↗

arXiv2016

Feature Learning for Chord Recognition: The Deep Chroma Extractor

We explore frame-level audio feature learning for chord recognition using artificial neural networks. We present the argument that chroma vectors potentially hold enough information to model harmonic content of audio for chord recognition, but that standard chroma extractors compute too noisy features. This leads us to propose a learned chroma feature extractor based on artificial neural networks. It is trained to compute chroma features that encode harmonic information important for chord recognition, while being robust to irrelevant interferences. We achieve this by feeding the network an audio spectrum with context instead of a single frame as input. This way, the network can learn to selectively compensate noise and resolve harmonic ambiguities. We compare the resulting features to hand-crafted ones by using a simple linear frame-wise classifier for chord recognition on various data sets. The results show that the learned feature extractor produces superior chroma vectors for chord recognition.

cs.SD↗