Single-channel Speech Enhancement Using Learnable Loss Mixup

Abstract

Generalization remains a major problem in supervised learning of single-channel speech enhancement. In this work, we propose learnable loss mixup (LLM), a simple and effortless training diagram, to improve the generalization of deep learning-based speech enhancement models. Loss mixup, of which learnable loss mixup is a special variant, optimizes a mixture of the loss functions of random sample pairs to train a model on virtual training data constructed from these pairs of samples. In learnable loss mixup, by conditioning on the mixed data, the loss functions are mixed using a non-linear mixing function automatically learned via neural parameterization. Our experimental results on the VCTK benchmark show that learnable loss mixup achieves 3.26 PESQ, outperforming the state-of-the-art.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keychang2023single

Related papers

• Mixup-breakdown: A Consistency Training Method For Improving Generalization Of Speech Separation Models (2019)0.00
• Unsupervised Speech Enhancement With Speech Recognition Embedding And Disentanglement Losses (2021)8.35
• Leveraging Heteroscedastic Uncertainty In Learning Complex Spectral Mapping For Single-channel Speech Enhancement (2022)3.58
• Self-supervised Learning From Contrastive Mixtures For Personalized Speech Enhancement (2020)0.00
• A Consolidated View Of Loss Functions For Supervised Deep Learning-based Speech Enhancement (2020)13.93
• Superm2m: Supervised And Mixture-to-mixture Co-learning For Speech Enhancement And Noise-robust ASR (2024)5.24
• Ctpulse: Close-talk, And Pseudo-label Based Far-field, Speech Enhancement (2024)0.00
• A Modulation-domain Loss For Neural-network-based Real-time Speech Enhancement (2021)8.09