---
title: "Learning Neural Vocoder from Range-Null Space Decomposition"
authors:
  - "Andong Li"
  - "Tong Lei"
  - "Zhihang Sun"
  - "Rilin Chen"
  - "Erwei Yin"
  - "Xiaodong Li"
  - "Chengshi Zheng"
doi: "10.24963/ijcai.2025/904"
url: "https://ijcai.org/proceedings/2025/904"
pdf: "https://ijcai.org/proceedings/2025/0904.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "8131-8140"
track: "Main Track"
keywords:
  - "Machine Learning: ML: Applications"
  - "Natural Language Processing: NLP: Tools"
  - "Natural Language Processing: NLP: Speech"
---

# Learning Neural Vocoder from Range-Null Space Decomposition

Andong Li, Tong Lei, Zhihang Sun, Rilin Chen, Erwei Yin, Xiaodong Li, Chengshi Zheng

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 8131-8140. DOI: [10.24963/ijcai.2025/904](https://doi.org/10.24963/ijcai.2025/904)

## Abstract

Despite the rapid development of neural vocoders in recent years, they usually suffer from some intrinsic challenges like opaque modeling, and parameter-performance trade-off. In this study, we propose an innovative time-frequency (T-F) domain-based neural vocoder to resolve the above-mentioned challenges. To be specific, we bridge the connection between the classical signal range-null decomposition (RND) theory and vocoder task, and the reconstruction of target spectrogram can be decomposed into the superimposition between the range-space and null-space, where the former is enabled by a linear domain shift from the original mel-scale domain to the target linear-scale domain, and the latter is instantiated via a learnable network for further spectral detail generation. Accordingly, we propose a novel dual-path framework, where the spectrum is hierarchically encoded/decoded, and the cross- and narrow-band modules are elaborately devised for efficient sub-band  and sequential modeling. Comprehensive experiments are conducted on the LJSpeech and LibriTTS benchmarks. Quantitative and qualitative results show that while enjoying lightweight network parameters, the proposed approach yields state-of-the-art performance among existing advanced methods. Our code and the pretrained model weights are available at https://github.com/Andong-Li-speech/RNDVoC.