Improved music transcription accuracy using Particle Filtering for PLCA model.
problem Limited performance of EM-based PLCA models in automatic music transcription.
method Employed Particle Filtering (PF) to overcome EM algorithm's limitations.
result Achieved 61.8% and 59.5% note-level transcription accuracy on two instrument repertoires.
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.
HMMs improve music transcription accuracy.
problem Improving automatic transcription of music.
method Employed PLCA for multi-pitch estimation and integrated HMMs for note segmentation and post-processing.
result HMMs enhance transcription accuracy on different instruments.
Scattering transform improves note onset detection and instrument recognition in music transcription.
problem Note onset detection and instrument recognition in music transcription.
method Multiscale scattering operators applied to MIDI-driven datasets and real musical pieces.
result Scattering transform outperforms other sound representations for note onset detection and instrument recognition.
Neural network generates music scores directly from polyphonic audio.
problem Transcribing music scores directly from polyphonic audio.
method Convolutional Recurrent Neural Network (CRNN) with CTC loss function.
result Model can learn to transcribe scores directly from audio signals.
Paper improves music transcription models with invariance and data augmentation.
problem Improving accuracy of frame-based music transcription models.
method Translation-invariant network combining filterbank and CNN, trained with pitch-shift augmented data.
result Top-performing model in MIREX evaluation, reducing model complexity and avoiding overfitting.
Real music signals are highly variable, yet they have strong statistical structure. Prior information about the underlying physical mechanisms by which sounds are generated and rules by which complex sound structure is constructed (notes, chords, a complete musical score), can be naturally unified using Bayesian modell…
We investigate the problem of modeling symbolic sequences of polyphonic music in a completely general piano-roll representation. We introduce a probabilistic model based on distribution estimators conditioned on a recurrent neural network that is able to discover temporal dependencies in high-dimensional sequences. Our…
A new model cleans vocal note event annotations in music.
problem Erroneous labels in music datasets.
method Contrastive learning to automatically create local deformations of likely correct labels.
result Transcription model accuracy improves with the proposed strategy.
New method improves music transcription by treating frequency distributions holistically.
problem Small frequency shifts and variations in sound timbre harm traditional fit measures.
method Optimal transportation and new holistic frequency distribution measure.
result Simplified note templates lead to faster, state-of-the-art performance.
Improved piano transcription by predicting onsets and frames together.
problem Polyphonic piano music transcription accuracy.
method Deep convolutional and recurrent neural network trained to predict pitch onsets and frames.
result Over 100% relative improvement in note F1 score on MAPS dataset.
The study improves pitch detection in polyphonic music by learning harmonic priors.
problem Challenges in transcribing polyphonic music due to overlapping harmonics.
method Introduced Gaussian process priors and used variational Bayes for inference.
result Learning priors that fit the frequency content of sound events improves pitch detection.
New system for automatic music emotion recognition considers multiple emotions simultaneously.
problem Automatic recognition of simultaneous and multiplicity of emotions in music.
method Comparison of multilabel and multiclass machine learning algorithms on the Emotify dataset.
result The Geneva Emotional Music Scale 9 is adopted for multilabel and multiclass classification of music emotions.
We present a supervised neural network model for polyphonic piano music transcription. The architecture of the proposed model is analogous to speech recognition systems and comprises an acoustic model and a music language model. The acoustic model is a neural network used for estimating the probabilities of pitches in …
Spectrogram-Channels U-Net separates sounds by treating each channel as a source's spectrogram.
problem Sound source separation in music information retrieval.
method Adapting U-Net to treat each channel of the output as a source's spectrogram, balancing volumes between sources.
result State-of-the-art performance on singing voice and multi-instrument separation.
Universal music translation network across instruments and genres.
problem Translating music across different instruments, genres, and styles.
method Multi-domain wavenet autoencoder with a shared encoder and disentangled latent space trained end-to-end on waveforms.
result Achieves convincing translations even from domains not seen during training.
New method synthesizes piano training data, improving transcription performance.
problem Lack of large piano datasets limits note onset transcription models.
method Synthesizes arbitrary training data, models piano dynamics, avoids disentanglement problem.
result Achieves good transcription performance on MAPS dataset and excellent generalization.
Combines symbolic and raw audio models for structured, realistic-sound music generation.
problem Lack of long-range dependencies in raw audio models and unstructured music.
method Uses a Long Short Term Memory network for melodic structure and WaveNet for raw audio generation with symbolic conditioning.
result Creates structured, realistic-sounding compositions using both symbolic and raw audio models.
Music genre classification is an essential tool for music information retrieval systems and it has been finding critical applications in various media platforms. Two important problems of the automatic music genre classification are feature extraction and classifier design. This paper investigates inter-genre similarit…
Paper proposes M2H-GAN to improve speech theme identification.
problem Limited ASR transcripts for speech theme identification.
method Uses M2H-GAN, a GAN-based approach, to generate TRS-like ASR transcripts.
result Improves speech theme identification performance close to human levels.
Music SketchNet generates missing measures in incomplete music pieces, guided by user input.
problem Generating missing measures in incomplete monophonic musical pieces.
method Introducing SketchVAE for factorized representation of rhythm and pitch, and two discriminative architectures for guided music completion.
result Our approach outperforms state-of-the-art models in both objective and subjective evaluations.
Paper proposes a bijective approach for signal/symbol translation using variational auto-encoders.
problem Extracting symbolic information from signals, especially in music, is challenging and non-generic.
method Turned into a density estimation task, using two variational auto-encoders with additive constraint.
result Bijective signal/symbol translation achieved, allowing both signal-to-symbol and symbol-to-signal inference.
MIDI-VAE models music dynamics and instrumentation for style transfer.
problem Modeling and transferring musical style across different instruments and dynamics.
method Variational Autoencoder (VAE) for polyphonic music modeling and style transfer.
result MIDI-VAE successfully transfers musical style between different genres and instruments.
Paper proposes LSTM for automatic music generation.
problem Generating music without human intervention.
method Recurrent Neural Networks (RNN), specifically LSTM.
result Model generates new melodies with high accuracy.
Inverse Drum Machine separates drum mixes using transcription and synthesis.
problem Separating individual drum tracks from mixed recordings.
method Analysis-by-synthesis framework combining deep learning and automatic transcription.
result Separation quality comparable to supervised methods requiring isolated stems.
End-to-end ASR error detection using audio-transcript entailment.
problem Detecting transcription errors in ASR systems to prevent error propagation.
method Proposes a novel end-to-end approach using audio-transcript entailment, with acoustic and linguistic encoders.
result Achieves CER of 26.2% on all transcription errors and 23% on medical errors specifically, improving by 12% and 15.4% respectively over a strong baseline.
The paper categorizes music emotions and improves music retrieval.
problem Inefficient music retrieval based on album information.
method Categorical emotion expression, Fisher's separation theorem, feature extraction, Support Vector Machines.
result Maximum separability occurs between relaxing and epic music parts.
CRAN extracts music highlights using attention and recurrent layers.
problem Extracting valuable music highlights from signals.
method Convolutional Recurrent Attention Networks (CRAN) with attention mechanism.
result CRAN outperforms three baseline methods in highlighting extraction.
Throughout music history, theorists have identified and documented interpretable rules that capture the decisions of composers. This paper asks, "Can a machine behave like a music theorist?" It presents MUS-ROVER, a self-learning system for automatically discovering rules from symbolic music. MUS-ROVER performs feature…
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.
A new neural network separates vocals from music accompaniment.
problem Separating vocals from music accompaniment in recordings.
method Self-attention convolutional neural network (CNN) with densely-connected blocks.
result 19.5% relative improvement in vocals separation.
Black-box adversarial examples improve ASR system accuracy.
problem Improving ASR system accuracy through targeted adversarial examples.
method Combining genetic algorithms and gradient estimation for black-box attacks.
result Achieved 89.25% targeted attack similarity with 94.6% audio file similarity.
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.
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.
The paper improves ASR accuracy using semi-supervised learning and dropout.
problem Improving ASR accuracy with limited labeled data.
method Training a seed model on limited labeled data, using dropout for uncertainty, and data selection for diversity.
result The approach significantly reduces ASR errors compared to baseline.
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.
Deep learning model estimates multiple f0s, melodies, vocals, and bass lines from music.
problem Estimating f0s and other musical elements from polyphonic music.
method Multitask deep learning architecture trained on a large dataset.
result Multitask model outperforms single-task models.
Word2Vec captures musical relationships in complex polyphonic music.
problem Capturing meaningful relationships in musical contexts.
method Skip-gram version of word2vec applied to music slices from a large corpus.
result Word2Vec embeddings reveal functional chord and harmonic associations.
Neural nets classify Thai Lukthung songs from other genres.
problem Classifying Thai Lukthung songs for personalized music recommendations.
method Developed neural networks using lyrics and audio features.
result Combined model yields F1 score of 0.86 for Lukthung classification. Improves seq2seq speech recognition by addressing overconfidence and incomplete transcriptions.
problem Overconfidence and incomplete transcriptions in seq2seq speech recognition.
method Proposed practical solutions to address overconfidence and incomplete transcriptions using a trigram language model.
result Achieved competitive speaker independent word error rates (6.7%) with a trigram language model.
Study combines speaker verification and voice trigger detection in a single network.
problem Separate training for speaker verification and voice trigger detection.
method Multi-task learning with a single network trained on both tasks.
result Single network achieves comparable accuracy to independent models for each task.
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.
Automated vowel duration measurement improves on manual methods.
problem Manual vowel duration annotation is time-consuming and inconsistent.
method Structured prediction framework; acoustic feature functions; learning algorithm to minimize duration error.
result Trained model outperforms HMM-based aligner in automatic duration estimation.
Spoken language translation (SLT) has become very important in an increasingly globalized world. Machine translation (MT) for automatic speech recognition (ASR) systems is a major challenge of great interest. This research investigates that automatic sentence segmentation of speech that is important for enriching speec…
Maximum Entropy model captures musical styles without plagiarizing them.
problem Generating music that sounds like a specific style without copying.
method A Maximum Entropy model with pairwise interactions, avoiding Markov interactions.
result Generated melodies capture the style of the original corpus without plagiarism.
New method predicts target genres from source genres, unifying music tag systems.
problem Automatic genre inference fails to handle music genre diversity and subjectivity.
method Knowledge-based, statistical, and hybrid translation models.
result Hybrid translation model is most effective for multilabel classification.
Survey on GANs for generating visual arts, music, and literature.
problem Tackles the challenge of generating art using GANs.
method Uses generative adversarial networks (GANs) to generate visual arts, music, and literary text.
result Performance comparison and description of various GAN architectures presented.
Paper improves multi-step chord prediction in jazz music.
problem Chord sequence prediction accuracy is poor in multi-step scenarios.
method Aggregated multi-scale encoder-decoder network architecture.
result Our model outperforms state-of-the-art methods in accuracy and perplexity.