# Sidon: Fast and Robust Open-Source Multilingual Speech Restoration for Large-scale Dataset Cleansing

arXiv: https://arxiv.org/abs/2509.17052 · HTML: https://mlsanity.com/paper/2509.17052/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Sep 2025.

## Built on by (most cited first, 8 of 8)

- [OmniVoice: Towards Omnilingual Zero-Shot Text-to-Speech with Diffusion Language Models](https://mlsanity.com/paper/2604.00688/) (2026)
- [VoXtream2: Full-stream TTS with dynamic speaking rate control](https://mlsanity.com/paper/2603.13518/) (2026)
- [Geneses: Unified Generative Speech Enhancement and Separation](https://mlsanity.com/paper/2601.18456/) (2026)
- [AnimeScore: A Preference-Based Dataset and Framework for Evaluating Anime-Like Speech Style](https://mlsanity.com/paper/2603.11482/) (2026)
- [What Counts as Real? Speech Restoration and Voice Quality Conversion Pose New Challenges to Deepfake Detection](https://mlsanity.com/paper/2603.14033/) (2026)
- [CleanCodec: Efficient and Robust Speech Tokenization via Perceptually Guided Encoding](https://mlsanity.com/paper/2606.04418/) (2026)
- [Controlling Backchannels in Streamable Full-duplex Models](https://mlsanity.com/paper/2609.29418/) (2026)
- [Learning Natural Conversational Behavior in Tandem Speech-to-Speech Models with Randomized Guidance](https://mlsanity.com/paper/2609.30773/) (2026)

## Abstract (arXiv)

Large-scale text-to-speech (TTS) systems are limited by the scarcity of clean, multilingual recordings. We introduce Sidon, a fast, open-source speech restoration model that converts noisy in-the-wild speech into studio-quality speech and scales to dozens of languages. Sidon consists of two models: w2v-BERT 2.0 finetuned feature predictor to cleanse features from noisy speech and vocoder trained to synthesize restored speech from the cleansed features. Sidon achieves restoration performance comparable to Miipher: Google's internal speech restoration model with the aim of dataset cleansing for speech synthesis. Sidon is also computationally efficient, running up to 500 times faster than real time on a single GPU. We further show that training a TTS model using a Sidon-cleansed automatic speech recognition corpus improves the quality of synthetic speech in a zero-shot setting. Code and model are released to facilitate reproducible dataset cleansing for the research community.