Research map / Image generation / Autoregressive and latent generation
PerformanceNet: Score-to-Audio Music Generation with Multi-Band Convolutional Residual Network
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use PerformanceNet instead of Synthesizer 2 (Logic Pro) | score-to-audio generation for violin, cello, and flute · User study, three instruments | naturalness MOS: 0.29 | Helped |
| Use PerformanceNet instead of either off-the-shelf synthesizer | score-to-audio generation for violin, cello, and flute · User study, three instruments | emotion MOS | Helped |
| Use PerformanceNet instead of Synthesizer 1 (Aria Maestosa) | score-to-audio generation for violin, cello, and flute · User study, three instruments | timbre MOS | Mixed |
| Use PerformanceNet instead of Synthesizer 2 (Logic Pro) | score-to-audio generation for violin, cello, and flute · User study, three instruments | audio quality MOS | Hurt |
| Use PerformanceNet instead of the WaveNet-based model | cello score-to-audio generation · User study, cello, two pianorolls | naturalness MOS | Helped |
| Use PerformanceNet instead of the WaveNet-based model | cello score-to-audio generation · User study, cello, two pianorolls | emotion MOS | Helped |
| Add TextureNet multi-band residual refinement (ablation) | ContourNet score-to-audio model · Score-to-audio generation | audio quality | Helped |
| Add the onset/offset encoder (ablation) | ContourNet score-to-audio model · Score-to-audio generation | note articulation | Helped |
| Use overlapping frequency bands and a Hamming window (ablation) | TextureNet · Score-to-audio generation | audio quality | No clear effect |
Closest papers by meaning
- MuseGAN: Multi-track Sequential Generative Adversarial Networks for Symbolic Music Generation and Accompaniment (AAAI 2018 · similarity 0.91)
- JEN-1 Composer: A Unified Framework for High-Fidelity Multi-Track Music Generation (AAAI 2025 · similarity 0.89)
- Deep Learning for Style Transfer and Experimentation with Audio Effects and Music Creation (AAAI 2024 · similarity 0.88)
- MusER: Musical Element-Based Regularization for Generating Symbolic Music with Emotion (AAAI 2024 · similarity 0.87)
- Whole-Song Hierarchical Generation of Symbolic Music Using Cascaded Diffusion Models (ICLR 2024 · similarity 0.87)
- LLark: A Multimodal Instruction-Following Language Model for Music (ICML 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.