Expressive music performance rendering involves interpreting symbolic scores with variations in timing, dynamics, articulation, and instrument-specific techniques, resulting in performances that capture musical can emotional intent. We introduce RenderBox, a unified framework for text-and-score controlled audio performance generation across multiple instruments, applying coarse-level controls through natural language descriptions and granular-level controls using music scores. Based on a diffusion transformer architecture and cross-attention joint conditioning, we propose a curriculum-based paradigm that trains from plain synthesis to expressive performance, gradually incorporating controllable factors such as speed, mistakes, and style diversity. RenderBox achieves high performance compared to baseline models across key metrics such as FAD and CLAP, and also tempo and pitch accuracy under different prompting tasks. Subjective evaluation further demonstrates that RenderBox is able to generate controllable expressive performances that sound natural and musically engaging, aligning well with prompts and intent.
In the pursuit of developing expressive music performance models using artificial intelligence, this paper introduces DExter, a new approach leveraging diffusion probabilistic models to render Western classical piano performances. In this approach, performance parameters are represented in a continuous expression space and a diffusion model is trained to predict these continuous parameters while being conditioned on the musical score. Furthermore, DExter also enables the generation of interpretations (expressive variations of a performance) guided by perceptually meaningful features by conditioning jointly on score and perceptual feature representations. Consequently, we find that our model is useful for learning expressive performance, generating perceptually steered performances, and transferring performance styles. We assess the model through quantitative and qualitative analyses, focusing on specific performance metrics regarding dimensions like asynchrony and articulation, as well as through listening tests comparing generated performances with different human interpretations. Results show that DExter is able to capture the time-varying correlation of the expressive parameters, and compares well to existing rendering models in subjectively evaluated ratings. The perceptual-feature-conditioned generation and transferring capabilities of DExter are verified by a proxy model predicting perceptual characteristics of differently steered performances.
Capturing and generating a wide variety of musical expression is important in music performance rendering, but current methods fail to model such a variation. This paper presents a music performance rendering method that could explicitly model differences in interpretations for a given piece of music. Conditional variational auto-encoder is used to jointly train, conditioned on the music score, an encoder from performance to a latent code and a decoder from the latent code to music performance. Evaluation demonstrates the method is capable of generating a wide variety of human-like expressive music performances as the latent code is varied.
In this paper, we present our application of deep neural network to modeling piano performance, which imitates the expressive control of tempo, dynamics, articulations and pedaling from pianists. Our model consists of recurrent neural networks with hierarchical attention and conditional variational autoencoder. The model takes a sequence of note-level score features extracted from MusicXML as input and predicts piano performance features of the corresponding notes. To render musical expressions consistently over long-term sections, we first predict tempo and dynamics in measure-level and, based on the result, refine them in note-level. The evaluation through listening test shows that our model achieves a more human-like expressiveness compared to previous models. We also share the dataset we used for the experiment.
Capturing intricate and subtle variations in human expressiveness in music performance using computational approaches is challenging. In this paper, we propose a novel approach for reconstructing human expressiveness in piano performance with a multi-layer bi-directional Transformer encoder. To address the needs for large amounts of accurately captured and score-aligned performance data in training neural networks, we use transcribed scores obtained from an existing transcription model to train our model. We integrate pianist identities to control the sampling process and explore the ability of our system to model variations in expressiveness for different pianists. The system is evaluated through statistical analysis of generated expressive performances and a listening test. Overall, the results suggest that our method achieves state-of-the-art in generating human-like piano performances from transcribed scores, while fully and consistently reconstructing human expressiveness poses further challenges.