# Training Flow Matching Models with Reliable Labels via Self-Purification

arXiv: https://arxiv.org/abs/2509.19091 · HTML: https://mlsanity.com/paper/2509.19091/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Audio, speech and music, one of 14 directions in Diffusion models.
- Builds on 22 of its 27 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Placed in

- Diffusion models / [Audio, speech and music](https://mlsanity.com/diffusion/audio-speech-and-music/)

## Builds on (most cited first, 22 of 22)

- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [VoxCeleb: a large-scale speaker identification dataset](https://mlsanity.com/paper/1706.08612/) (2017)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speech](https://mlsanity.com/paper/2105.06337/) (2021)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [WhisperX: Time-Accurate Speech Transcription of Long-Form Audio](https://mlsanity.com/paper/2303.00747/) (2023)
- [DNSMOS: A Non-Intrusive Perceptual Objective Speech Quality metric to evaluate Noise Suppressors](https://mlsanity.com/paper/2010.15258/) (2020)
- [Emilia: An Extensive, Multilingual, and Diverse Speech Dataset for Large-Scale Speech Generation](https://mlsanity.com/paper/2407.05361/) (2024)
- [Real Time Speech Enhancement in the Waveform Domain](https://mlsanity.com/paper/2006.12847/) (2020)
- [Robustness of Conditional GANs to Noisy Labels](https://mlsanity.com/paper/1811.03205/) (2018)
- [ESPnet-Codec: Comprehensive Training and Evaluation of Neural Codecs for Audio, Music, and Speech](https://mlsanity.com/paper/2409.15897/) (2024)
- [Label-Noise Robust Generative Adversarial Networks](https://mlsanity.com/paper/1811.11165/) (2018)
- [A Vector Quantized Approach for Text to Speech Synthesis on Real-World Spontaneous Speech](https://mlsanity.com/paper/2302.04215/) (2023)
- [Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels](https://mlsanity.com/paper/2305.19518/) (2023)
- [Label-Noise Robust Diffusion Models](https://mlsanity.com/paper/2402.17517/) (2024)
- [Text-To-Speech Synthesis In The Wild](https://mlsanity.com/paper/2409.08711/) (2024)
- [Guiding Noisy Label Conditional Diffusion Models with Score-based Discriminator Correction](https://mlsanity.com/paper/2508.19581/) (2025)
- [Length-Aware Rotary Position Embedding for Text-Speech Alignment](https://mlsanity.com/paper/2509.11084/) (2025)

## Built on by (most cited first, 1 of 1)

- [Robust TTS Training via Self-Purifying Flow Matching for the WildSpoof 2026 TTS Track](https://mlsanity.com/paper/2512.17293/) (2025)

## Abstract (arXiv)

Training datasets are inherently imperfect, often containing mislabeled samples due to human annotation errors, limitations of tagging models, and other sources of noise. Such label contamination can significantly degrade the performance of a trained model. In this work, we introduce Self-Purifying Flow Matching (SPFM), a principled approach to filtering unreliable data within the flow-matching framework. SPFM identifies suspicious data using the model itself during the training process, bypassing the need for pretrained models or additional modules. Our experiments demonstrate that models trained with SPFM generate samples that accurately adhere to the specified conditioning, even when trained on noisy labels. Furthermore, we validate the robustness of SPFM on the TITW dataset, which consists of in-the-wild speech data, achieving performance that surpasses existing baselines.