The paper tests deep music embeddings for semantic consistency.
problem Ensuring deep music embeddings capture meaningful musical semantics.
method Proposes a systematic method to test deep music representations for semantic consistency, considering both input audio space and latent deep space.
result Distance consistency between related points is maintained in both input audio space and latent deep space.
Isolating individual instruments in a musical mixture has a myriad of potential applications, and seems imminently achievable given the levels of performance reached by recent deep learning methods. While most musical source separation techniques learn an independent model for each instrument, we propose using a common…
Trains word embeddings from music and text data to link music contexts.
problem Varying vocabulary size and musical relevance in word embeddings.
method Combines general text and music-specific data to train word embeddings.
result Trained embeddings better associate music contexts with compositions.
We explore the potential of a popular distributional semantics vector space model, word2vec, for capturing meaningful relationships in ecological (complex polyphonic) music. More precisely, the skip-gram version of word2vec is used to model slices of music from a large corpus spanning eight musical genres. In this newl…
MuLan links music audio to natural language tags.
problem Traditional music tagging systems use rigid attributes; MuLan aims to link audio directly to natural language.
method Joint audio-text embedding model trained on 44 million music recordings and text annotations.
result MuLan's embeddings enable zero-shot functionalities and transfer learning.
Proposes COALA method for learning audio representations aligned with tags.
problem Lack of annotated data for high-performance audio representation learning.
method Aligns latent representations of audio and tags using a contrastive loss.
result Audio embedding model captures both acoustic and semantic characteristics.
We propose a novel approach for the generation of polyphonic music based on LSTMs. We generate music in two steps. First, a chord LSTM predicts a chord progression based on a chord embedding. A second LSTM then generates polyphonic music from the predicted chord progression. The generated music sounds pleasing and harm…
Deep clustering is the first method to handle general audio separation scenarios with multiple sources of the same type and an arbitrary number of sources, performing impressively in speaker-independent speech separation tasks. However, little is known about its effectiveness in other challenging situations such as mus…
Improved music recommendations using hyperbolic embeddings and Bayesian estimation.
problem Generating personalized music recommendations.
method Hyperbolic Poincaré embeddings for hierarchies, empirical Bayes for link reliability.
result Significant performance improvement in A/B testing.
A model of music needs to have the ability to recall past details and have a clear, coherent understanding of musical structure. Detailed in the paper is a deep reinforcement learning architecture that predicts and generates polyphonic music aligned with musical rules. The probabilistic model presented is a Bi-axial LS…
Music generation research has grown in popularity over the past decade, thanks to the deep learning revolution that has redefined the landscape of artificial intelligence. In this paper, we propose a novel approach to music generation inspired by musical segment concatenation methods and hash learning algorithms. Given…
Model generates music to connect missing parts, leveraging latent space of VAE.
problem Music inpainting from missing or lost information.
method Deep learning model using VAE latent space and RNN to traverse latent space conditioned on past and future musical contexts.
result Model generates meaningful music inpaintings, connecting musical excerpts.
This paper explores deep learning in music generation, from history to current techniques.
problem Creating music automatically using deep learning.
method Analysis of historical and recent deep learning music generation systems.
result Deep learning can learn musical styles and generate music samples.
Dual-track model generates classical piano music better than other methods.
problem Generating inter-dependent piano music.
method Dual-track architecture with various neural network models and music representations.
result Proposed model outperforms other methods in evaluation.
As deep learning advances, algorithms of music composition increase in performance. However, most of the successful models are designed for specific musical structures. Here, we present BachProp, an algorithmic composer that can generate music scores in many styles given sufficient training data. To adapt BachProp to a…
Bi-LSTM with attention generates jazz music with rich nuances.
problem Challenges in producing realistic music with structure and rationality.
method Deep learning, Bi-directional Long Short Term Memory (Bi-LSTM) Neural Network with Attention.
result Bi-LSTMs with attention preserve the richness and technical nuances of old style music.
Aug-Gen improves music generation by augmenting training data with model-produced examples.
problem Lack of high-quality training data for specific musical styles.
method Dataset augmentation using generated examples during training.
result Aug-Gen improves both training duration and quality of generated music.
Low-dimensional vectors improve semantic understanding of music and language.
problem Noise in shared semantics due to individual brain biases.
method Jointly model multiple brains to learn low-dimensional vector embeddings.
result These embeddings outperform high-dimensional fMRI data in music and language classification.
Transformer autoencoder learns musical style from performances.
problem Learning high-level controls over symbolic music generation.
method Aggregates encodings of input data across time to obtain global style representation.
result Improves control over performance style and melody in music generation tasks.
The recent success of raw audio waveform synthesis models like WaveNet motivates a new approach for music synthesis, in which the entire process --- creating audio samples from a score and instrument information --- is modeled using generative neural networks. This paper describes a neural music synthesis model with fl…
Paper classifies Brazilian music genres using song lyrics with BLSTM network.
problem Classifying Brazilian music genres from lyrics.
method Used BLSTM network combined with SVM, Random Forest, and word embeddings.
result BLSTM outperforms other models with an F1-score of 0.48.
Grading function evaluates Bach-style chorales, outperforming human experts.
problem Difficulty in automatically evaluating musical style correctness.
method Introduces a grading function for evaluating four-part chorales in the style of J.S. Bach.
result Transformer model output is outperformed by the grading function at discriminating Bach chorales.
Graph neural networks improve music genre classification on audio datasets.
problem Difficulty in applying deep learning on spectrograms due to lack of quality data and augmentation.
method Combination of CNN and Graph Neural Networks (GNN) with Siamese Neural Networks.
result Achieved state-of-the-art results on GTZAN and AudioSet datasets.
FIGARO generates symbolic music with fine-grained control.
problem Minimal control over generated music sequences.
method Description-to-sequence task, learning conditional distribution of sequences given high-level descriptions.
result State-of-the-art controllable symbolic music generation.
New method separates music vocals from accompaniment without labeled data.
problem Separating music sources without isolated recordings.
method Bootstrapping deep model using primitive auditory cues.
result Trained deep model separates vocals from accompaniment in unlabeled music.
Transformer improves pop piano composition by incorporating beat-based structure.
problem Generating expressive pop piano compositions with coherent rhythmic structure.
method Improved data representation for Transformers, incorporating beat-bar-phrase structure.
result Composes pop piano music with better rhythmic structure than existing models.
Diffusion models generate music sequences without autoregressive loops.
problem Generating music sequences from symbolic data using diffusion models.
method Parameterize discrete symbolic data in continuous latent space, train diffusion model, generate sequences through reverse process.
result Strong unconditional generation and post-hoc conditional infilling compared to autoregressive models.
Generating music medleys is about finding an optimal permutation of a given set of music clips. Toward this goal, we propose a self-supervised learning task, called the music puzzle game, to train neural network models to learn the sequential patterns in music. In essence, such a game requires machines to correctly sor…
Review of deep learning techniques for audio signal processing.
problem Improving audio signal processing using deep learning.
method Analysis of various deep learning models and techniques.
result Advancements in speech, music, and environmental sound processing.
Feature learning and deep learning have drawn great attention in recent years as a way of transforming input data into more effective representations using learning algorithms. Such interest has grown in the area of music information retrieval (MIR) as well, particularly in music audio classification tasks such as auto…
Existing automatic music generation approaches that feature deep learning can be broadly classified into two types: raw audio models and symbolic models. Symbolic models, which train and generate at the note level, are currently the more prevalent approach; these models can capture long-range dependencies of melodic st…
Improved cover detection in music datasets with novel triplet loss.
problem Challenging task of automatically detecting covers in audio datasets.
method Convolutional neural network mapping melodic features to embeddings, training to minimize cover distance and maximize non-cover distance.
result New prototypical triplet loss improves accuracy for large datasets and live songs.
Adversarial learning improves music transcription accuracy.
problem Conditional independence of labels in deep learning models limits transcription performance.
method Adversarial training scheme operating on time-frequency representations to reduce inter-label dependencies.
result Adversarial learning reduces error rate and increases model confidence.
Bayesian sparsification improves complex-valued neural networks by 50-100x with minimal performance loss.
problem Efficiently compressing complex-valued neural networks for embedded systems.
method Extending Sparse Variational Dropout to complex-valued networks and conducting a numerical study.
result Achieved state-of-the-art performance on MusicNet with 50-100x compression.
The paper examines the consistency of item embeddings in recommendation systems.
problem The relevance of averaging item embeddings for user or concept representation.
method Proposes an expected precision score to measure consistency and analyzes it theoretically and empirically.
result Real-world averages are less consistent for recommendation compared to theoretical assumptions.
Improved music source separation using unlabeled data remixing.
problem Music source separation with deep learning.
method Introduces a simple convolutional and recurrent model and a scheme to leverage unlabeled music.
result Waveform methods can now match spectrogram methods on standard benchmarks.
Generative model blends query input with latent states for structured improvisation.
problem Generating structured music from latent states of a neural network.
method Used a Variational Autoencoder (VAE) trained on a specific style corpus, and controlled blending with a noisy channel.
result Generated music with longer-term structure that blends query input with network style.
In the domain of algorithmic music composition, machine learning-driven systems eliminate the need for carefully hand-crafting rules for composition. In particular, the capability of recurrent neural networks to learn complex temporal patterns lends itself well to the musical domain. Promising results have been observe…
Deep generative models such as Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) have recently been applied to style and domain transfer for images, and in the case of VAEs, music. GAN-based models employing several generators and some form of cycle consistency loss have been among the most suc…
Paper prunes deep MIR models to ultra-light versions.
problem Massive complexity of deep learning models in MIR.
method Lottery ticket hypothesis-based model pruning.
result Up to 90% of model parameters can be removed without loss of accuracy.
EC^2-VAE generates music analogies by disentangling pitch and rhythm representations.
problem Disentangling music representations for generating creative analogies.
method Explicitly-constrained variational autoencoder (EC^2-VAE) for disentangling pitch and rhythm representations.
result EC^2-VAE enables the generation of music analogies by borrowing representations from different pieces.
Study develops supervised algorithm for symbolic music style translation.
problem Transforming musical style while maintaining original content.
method Synthetic data generation for aligned training data, encoder-decoder model.
result Trained models produce meaningful accompaniments for real MIDI recordings.
New system tracks musical performances in raw sheet images without preprocessing.
problem Lack of direct score position estimation in raw sheet images.
method Proposes an Audio-Conditioned U-Net architecture.
result Direct score position estimation in entire unprocessed sheet images.
Speech, Music and Noise classification/segmentation is an important preprocessing step for audio processing/indexing. To this end, we propose a novel 1D Convolutional Neural Network (CNN) - SwishNet. It is a fast and lightweight architecture that operates on MFCC features which is suitable to be added to the front-end …
New model explains music emotion predictions with visualizable mid-level features.
problem Challenging to quantify and predict music emotion.
method VGG-style deep neural network with mid-level perceptual features.
result Small loss in performance justifies high explainability of predictions.
Synthesizing human's movements such as dancing is a flourishing research field which has several applications in computer graphics. Recent studies have demonstrated the advantages of deep neural networks (DNNs) for achieving remarkable performance in motion and music tasks with little effort for feature pre-processing.…
Deep learning models generate music with arbitrary control strategies.
problem Lack of efficient methods for generating music with arbitrary control.
method Deep generative models learn to navigate arbitrary sound spaces.
result Deep learning enables high-quality, arbitrary sound synthesis.
Generates music with video emotion using deep neural networks.
problem Generating music with video emotion.
method Hybrid deep neural network combining ANFIS and LSTM.
result Low mean absolute errors and similar global features in spectrograms.