This paper explores the idea of utilising Long Short-Term Memory neural networks (LSTMNN) for the generation of musical sequences in ABC notation. The proposed approach takes ABC notations from the Nottingham dataset and encodes it to beefed as input for the neural networks. The primary objective is to input the neural networks with an arbitrary note, let the network process and augment a sequence based on the note until a good piece of music is produced. Multiple tunings have been done to amend the parameters of the network for optimal generation. The output is assessed on the basis of rhythm, harmony, and grammar accuracy.
While deep neural network-based music source separation (MSS) is very effective and achieves high performance, its model size is often a problem for practical deployment. Deep implicit architectures such as deep equilibrium models (DEQ) were recently proposed, which can achieve higher performance than their explicit counterparts with limited depth while keeping the number of parameters small. This makes DEQ also attractive for MSS, especially as it was originally applied to sequential modeling tasks in natural language processing and thus should in principle be also suited for MSS. However, an investigation of a good architecture and training scheme for MSS with DEQ is needed as the characteristics of acoustic signals are different from those of natural language data. Hence, in this paper we propose an architecture and training scheme for MSS with DEQ. Starting with the architecture of Open-Unmix (UMX), we replace its sequence model with DEQ. We refer to our proposed method as DEQ-based UMX (DEQ-UMX). Experimental results show that DEQ-UMX performs better than the original UMX while reducing its number of parameters by 30%.
We demonstrate that language models pre-trained on codified (discretely-encoded) music audio learn representations that are useful for downstream MIR tasks. Specifically, we explore representations from Jukebox (Dhariwal et al. 2020): a music generation system containing a language model trained on codified audio from 1M songs. To determine if Jukebox's representations contain useful information for MIR, we use them as input features to train shallow models on several MIR tasks. Relative to representations from conventional MIR models which are pre-trained on tagging, we find that using representations from Jukebox as input features yields 30% stronger performance on average across four MIR tasks: tagging, genre classification, emotion recognition, and key detection. For key detection, we observe that representations from Jukebox are considerably stronger than those from models pre-trained on tagging, suggesting that pre-training via codified audio language modeling may address blind spots in conventional approaches. We interpret the strength of Jukebox's representations as evidence that modeling audio instead of tags provides richer representations for MIR.
This paper presents CQT-Diff, a data-driven generative audio model that can, once trained, be used for solving various different audio inverse problems in a problem-agnostic setting. CQT-Diff is a neural diffusion model with an architecture that is carefully constructed to exploit pitch-equivariant symmetries in music. This is achieved by preconditioning the model with an invertible Constant-Q Transform (CQT), whose logarithmically-spaced frequency axis represents pitch equivariance as translation equivariance. The proposed method is evaluated with objective and subjective metrics in three different and varied tasks: audio bandwidth extension, inpainting, and declipping. The results show that CQT-Diff outperforms the compared baselines and ablations in audio bandwidth extension and, without retraining, delivers competitive performance against modern baselines in audio inpainting and declipping. This work represents the first diffusion-based general framework for solving inverse problems in audio processing.
We propose a new task named Audio-driven Per-formance Video Generation (APVG), which aims to synthesizethe video of a person playing a certain instrument guided bya given music audio clip. It is a challenging task to gener-ate the high-dimensional temporal consistent videos from low-dimensional audio modality. In this paper, we propose a multi-staged framework to achieve this new task to generate realisticand synchronized performance video from given music. Firstly,we provide both global appearance and local spatial informationby generating the coarse videos and keypoints of body and handsfrom a given music respectively. Then, we propose to transformthe generated keypoints to heatmap via a differentiable spacetransformer, since the heatmap offers more spatial informationbut is harder to generate directly from audio. Finally, wepropose a Structured Temporal UNet (STU) to extract bothintra-frame structured information and inter-frame temporalconsistency. They are obtained via graph-based structure module,and CNN-GRU based high-level temporal module respectively forfinal video generation. Comprehensive experiments validate theeffectiveness of our proposed framework.
Text-to-speech (TTS) and singing voice synthesis (SVS) aim at generating high-quality speaking and singing voice according to textual input and music scores, respectively. Unifying TTS and SVS into a single system is crucial to the applications requiring both of them. Existing methods usually suffer from some limitations, which rely on either both singing and speaking data from the same person or cascaded models of multiple tasks. To address these problems, a simplified elegant framework for TTS and SVS, named UniSyn, is proposed in this paper. It is an end-to-end unified model that can make a voice speak and sing with only singing or speaking data from this person. To be specific, a multi-conditional variational autoencoder (MC-VAE), which constructs two independent latent sub-spaces with the speaker- and style-related (i.e. speak or sing) conditions for flexible control, is proposed in UniSyn. Moreover, supervised guided-VAE and timbre perturbation with the Wasserstein distance constraint are leveraged to further disentangle the speaker timbre and style. Experiments conducted on two speakers and two singers demonstrate that UniSyn can generate natural speaking and singing voice without corresponding training data. The proposed approach outperforms the state-of-the-art end-to-end voice generation work, which proves the effectiveness and advantages of UniSyn.
Musical Metacreation tries to obtain creative behaviors from computers algorithms composing music. In this paper I briefly analyze how this field evolved from algorithmic composition to be focused on the search for creativity, and I point out some issues in pursuing this goal. Finally, I argue that hybridization of algorithms can be a useful direction for research.
With the ever-increasing number of digital music and vast music track features through popular online music streaming software and apps, feature recognition using the neural network is being used for experimentation to produce a wide range of results across a variety of experiments recently. Through this work, the authors extract information on such features from a popular open-source music corpus and explored new recognition techniques, by applying unsupervised Hebbian learning techniques on their single-layer neural network using the same dataset. The authors show the detailed empirical findings to simulate how such an algorithm can help a single layer feedforward network in training for music feature learning as patterns. The unsupervised training algorithm enhances their proposed neural network to achieve an accuracy of 90.36% for successful music feature detection. For comparative analysis against similar tasks, authors put their results with the likes of several previous benchmark works. They further discuss the limitations and thorough error analysis of their work. The authors hope to discover and gather new information about this particular classification technique and its performance, and further understand future potential directions and prospects that could improve the art of computational music feature recognition.
Optical Music Recognition is a field that attempts to extract digital information from images of either the printed music scores or the handwritten music scores. One of the challenges of the Optical Music Recognition task is to transcript the symbols of the camera-captured images into digital music notations. Previous end-to-end model, based on deep learning, was developed as a Convolutional Recurrent Neural Network. However, it does not explore sufficient contextual information from full scales and there is still a large room for improvement. In this paper, we propose an innovative end-to-end framework that combines a block of Residual Recurrent Convolutional Neural Network with a recurrent Encoder-Decoder network to map a sequence of monophonic music symbols corresponding to the notations present in the image. The Residual Recurrent Convolutional block can improve the ability of the model to enrich the context information while the number of parameter will not be increasing. The experiment results were benchmarked against a publicly available dataset called CAMERA-PRIMUS. We evaluate the performances of our model on both the images with ideal conditions and that with non-ideal conditions. The experiments show that our approach surpass the state-of-the-art end-to-end method using Convolutional Recurrent Neural Network.
We propose a unified model for three inter-related tasks: 1) to \textit{separate} individual sound sources from a mixed music audio, 2) to \textit{transcribe} each sound source to MIDI notes, and 3) to\textit{ synthesize} new pieces based on the timbre of separated sources. The model is inspired by the fact that when humans listen to music, our minds can not only separate the sounds of different instruments, but also at the same time perceive high-level representations such as score and timbre. To mirror such capability computationally, we designed a pitch-timbre disentanglement module based on a popular encoder-decoder neural architecture for source separation. The key inductive biases are vector-quantization for pitch representation and pitch-transformation invariant for timbre representation. In addition, we adopted a query-by-example method to achieve \textit{zero-shot} learning, i.e., the model is capable of doing source separation, transcription, and synthesis for \textit{unseen} instruments. The current design focuses on audio mixtures of two monophonic instruments. Experimental results show that our model outperforms existing multi-task baselines, and the transcribed score serves as a powerful auxiliary for separation tasks.