DiT-Flow: Speech Enhancement Robust to Multiple Distortions based on Flow Matching in Latent Space and Diffusion Transformers

• URL: http://arxiv.org/abs/2603.21608v1
• Date: Mon, 23 Mar 2026 06:03:58 GMT
• Title: DiT-Flow: Speech Enhancement Robust to Multiple Distortions based on Flow Matching in Latent Space and Diffusion Transformers
• Abstract summary: DiT-Flow is a flow matching-based SE framework built on the latent Diffusion Transformer (DiT) backbone.<n>We show that DiT-Flow consistently outperforms state-of-the-art generative SE models.
• Score: 22.00618863958246
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in generative models, such as diffusion and flow matching, have shown strong performance in audio tasks. However, speech enhancement (SE) models are typically trained on limited datasets and evaluated under narrow conditions, limiting real-world applicability. To address this, we propose DiT-Flow, a flow matching-based SE framework built on the latent Diffusion Transformer (DiT) backbone and trained for robustness across diverse distortions, including noise, reverberation, and compression. DiT-Flow operates on compact variational auto-encoders (VAEs)-derived latent features. We validated our approach on StillSonicSet, a synthetic yet acoustically realistic dataset composed of LibriSpeech, FSD50K, FMA, and 90 Matterport3D scenes. Experiments show that DiT-Flow consistently outperforms state-of-the-art generative SE models, demonstrating the effectiveness of flow matching in multi-condition speech enhancement. Despite ongoing efforts to expand synthetic data realism, a persistent bottleneck in SE is the inevitable mismatch between training and deployment conditions. By integrating LoRA with the MoE framework, we achieve both parameter-efficient and high-performance training for DiT-Flow robust to multiple distortions with using 4.9% percentage of the total parameters to obtain a better performance on five unseen distortions.

Related papers

• AudioNoisePrints: Model-free audio watermarking using spatial correlation in flow matching TTS [11.455535872504749]
  We present AudioNoisePrints, a training-free watermarking pipeline for flow matching and diffusion TTS models.<n>We exploited the fact that there are strong correlations between the initial Gaussian noises and the generated outputs in diffusion and flow matching models.<n>Our method outperforms AudioSeal, a strong baseline for audio watermarking under strong augmentations.
  arXiv  Detail & Related papers  (2026-08-23T02:52:37Z)
• dots.tts Technical Report [21.076643922579517]
  dots.tts is a continuous autoregressive text-to-speech foundation model.<n>We train an AudioVAE with multiple objectives to build a semantically structured and prediction-friendly continuous speech space.<n>We apply reward-free self-corrective post-training to the flow-matching head to further improve robustness and acoustic quality.
  arXiv  Detail & Related papers  (2026-06-05T09:19:24Z)
• Efficient Hybrid SE(3)-Equivariant Visuomotor Flow Policy via Spherical Harmonics for Robot Manipulation [51.48086451600577]
  E3Flow is a novel framework that addresses the limitations of equivariant diffusion policies.<n>It unifies efficient rectified flow with stable, multi-modal equivariant learning for the first time.<n>E3Flow achieves a 3.12% improvement in average success rate over the state-of-the-art Spherical Diffusion Policy.
  arXiv  Detail & Related papers  (2026-03-24T14:00:36Z)
• High-Performance Self-Supervised Learning by Joint Training of Flow Matching [1.8659515282266286]
  Flow Matching-based Foundation Model (FlowFM)<n>FlowFM reduces training time by 50.4% compared to a diffusion-based approach.<n>On downstream tasks, FlowFM surpassed the state-of-the-art SSL method (SSL-Wearables) on all five datasets.
  arXiv  Detail & Related papers  (2025-12-17T06:35:03Z)
• MeanFlow-Accelerated Multimodal Video-to-Audio Synthesis via One-Step Generation [12.665130073406651]
  A key challenge in synthesizing audios from silent videos is the inherent trade-off between synthesis quality and inference efficiency.<n>We introduce a MeanFlow-accelerated model that characterizes flow fields using average velocity.<n>We demonstrate that incorporating MeanFlow into the network significantly improves inference speed without compromising perceptual quality.
  arXiv  Detail & Related papers  (2025-09-08T07:15:21Z)
• Temporal In-Context Fine-Tuning for Versatile Control of Video Diffusion Models [34.131515004434846]
  We introduce Temporal In-Context Fine-Tuning (TIC-FT), an efficient approach for adapting pretrained video diffusion models to conditional generation tasks.<n>TIC-FT requires no architectural changes and achieves strong performance with as few as 10-30 training samples.<n>We validate our method across a range of tasks, including image-to-video and video-to-video generation, using large-scale base models such as CogVideoX-5B and Wan-14B.
  arXiv  Detail & Related papers  (2025-06-01T12:57:43Z)
• Communication-Efficient Wireless Federated Fine-Tuning for Large-Scale AI Models [13.742950928229078]
  Low-Rank Adaptation (LoRA) addresses these issues by training compact, low-rank matrices instead of fully fine-tuning large models.<n>This paper introduces a wireless federated LoRA fine-tuning framework that optimize both learning performance and communication efficiency.
  arXiv  Detail & Related papers  (2025-05-01T06:15:38Z)
• One-Step Diffusion Model for Image Motion-Deblurring [85.76149042561507]
  We propose a one-step diffusion model for deblurring (OSDD), a novel framework that reduces the denoising process to a single step.<n>To tackle fidelity loss in diffusion models, we introduce an enhanced variational autoencoder (eVAE), which improves structural restoration.<n>Our method achieves strong performance on both full and no-reference metrics.
  arXiv  Detail & Related papers  (2025-03-09T09:39:57Z)
• FlowTS: Time Series Generation via Rectified Flow [67.41208519939626]
  FlowTS is an ODE-based model that leverages rectified flow with straight-line transport in probability space.<n>For unconditional setting, FlowTS achieves state-of-the-art performance, with context FID scores of 0.019 and 0.011 on Stock and ETTh datasets.<n>For conditional setting, we have achieved superior performance in solar forecasting.
  arXiv  Detail & Related papers  (2024-11-12T03:03:23Z)
• DiTAS: Quantizing Diffusion Transformers via Enhanced Activation Smoothing [5.174900115018253]
  We propose a data-free post-training quantization (PTQ) method for efficient Diffusion Transformers (DiTs) DiTAS relies on the proposed temporal-aggregated smoothing techniques to mitigate the impact of the channel-wise outliers within the input activations. We show that our approach enables 4-bit weight, 8-bit activation (W4A8) quantization for DiTs while maintaining comparable performance as the full-precision model.
  arXiv  Detail & Related papers  (2024-09-12T05:18:57Z)
• Boundary-aware Decoupled Flow Networks for Realistic Extreme Rescaling [49.215957313126324]
  Recently developed generative methods, including invertible rescaling network (IRN) based and generative adversarial network (GAN) based methods, have demonstrated exceptional performance in image rescaling. However, IRN-based methods tend to produce over-smoothed results, while GAN-based methods easily generate fake details. We propose Boundary-aware Decoupled Flow Networks (BDFlow) to generate realistic and visually pleasing results.
  arXiv  Detail & Related papers  (2024-05-05T14:05:33Z)
• From PEFT to DEFT: Parameter Efficient Finetuning for Reducing Activation Density in Transformers [52.199303258423306]
  We propose a novel density loss that encourages higher activation sparsity in pre-trained models. Our proposed method, textbfDEFT, can consistently reduce activation density by up to textbf44.94% on RoBERTa$_mathrmLarge$ and by textbf53.19% (encoder density) and textbf90.60% (decoder density) on Flan-T5$_mathrmXXL$.
  arXiv  Detail & Related papers  (2024-02-02T21:25:46Z)
• High-Fidelity Speech Synthesis with Minimal Supervision: All Using Diffusion Models [56.00939852727501]
  Minimally-supervised speech synthesis decouples TTS by combining two types of discrete speech representations. Non-autoregressive framework enhances controllability, and duration diffusion model enables diversified prosodic expression.
  arXiv  Detail & Related papers  (2023-09-27T09:27:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.