---
title: "BridgeVoC: Neural Vocoder with Schrödinger Bridge"
authors:
  - "Tong Lei"
  - "Zhiyu Zhang"
  - "Rilin Chen"
  - "Meng Yu"
  - "Jing Lu"
  - "Chengshi Zheng"
  - "Dong Yu"
  - "Andong Li"
doi: "10.24963/ijcai.2025/903"
url: "https://ijcai.org/proceedings/2025/903"
pdf: "https://ijcai.org/proceedings/2025/0903.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "8122-8130"
track: "Main Track"
keywords:
  - "Natural Language Processing: NLP: Speech"
  - "Natural Language Processing: NLP: Applications"
  - "Natural Language Processing: NLP: Information extraction"
---

# BridgeVoC: Neural Vocoder with Schrödinger Bridge

Tong Lei, Zhiyu Zhang, Rilin Chen, Meng Yu, Jing Lu, Chengshi Zheng, Dong Yu, Andong Li

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 8122-8130. DOI: [10.24963/ijcai.2025/903](https://doi.org/10.24963/ijcai.2025/903)

## Abstract

While previous diffusion-based neural vocoders typically follow a noise-to-data generation pipe-line, the linear-degradation prior of the mel-spectrogram is often neglected, resulting in limited generation quality. By revisiting the vocoding task and excavating its connection with the signal restoration task, this paper proposes a time-frequency (T-F) domain-based neural vocoder with the Schrödinger Bridge, called BridgeVoC, which is the first to follow the data-to-data generation paradigm. Specifically, the mel-spectrogram can be projected into the target linear-scale domain and regarded as a degraded spectral representation with a deficient rank distribution. Based on this, the Schrödinger Bridge is leveraged to establish a connection between the degraded and target data distributions. During the inference stage, starting from the degraded representation, the target spectrum can be gradually restored rather than generated from a Gaussian noise process. Quantitative experiments on LJSpeech and LibriTTS show that BridgeVoC achieves faster inference and surpasses existing diffusion-based vocoder baselines, while also matching or exceeding non-diffusion state-of-the-art methods across evaluation metrics.