# Wave-U-Mamba: An End-To-End Framework For High-Quality And Efficient Speech Super Resolution

arXiv: https://arxiv.org/abs/2409.09337 · HTML: https://mlsanity.com/paper/2409.09337/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Speech and audio SSMs, one of 10 directions in State-space sequence models (S4, Mamba).
- Builds on 32 of its 40 references that are in the map.
- First posted on arXiv 14 Sep 2024.

## Placed in

- State-space sequence models (S4, Mamba) / [Speech and audio SSMs](https://mlsanity.com/ssm/speech-and-audio-ssms/)

## Builds on (most cited first, 30 of 32)

- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [WaveNet: A Generative Model for Raw Audio](https://mlsanity.com/paper/1609.03499/) (2016)
- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model](https://mlsanity.com/paper/2401.09417/) (2024)
- [Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions](https://mlsanity.com/paper/1712.05884/) (2017)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks](https://mlsanity.com/paper/1602.07868/) (2016)
- [Tacotron: Towards End-to-End Speech Synthesis](https://mlsanity.com/paper/1703.10135/) (2017)
- [VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text](https://mlsanity.com/paper/2104.11178/) (2021)
- [U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation](https://mlsanity.com/paper/2401.04722/) (2024)
- [Jamba: A Hybrid Transformer-Mamba Language Model](https://mlsanity.com/paper/2403.19887/) (2024)
- [Early Convolutions Help Transformers See Better](https://mlsanity.com/paper/2106.14881/) (2021)
- [MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis](https://mlsanity.com/paper/1910.06711/) (2019)
- [Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram](https://mlsanity.com/paper/1910.11480/) (2019)
- [VM-UNet: Vision Mamba UNet for Medical Image Segmentation](https://mlsanity.com/paper/2402.02491/) (2024)
- [It's Raw! Audio Generation with State-Space Models](https://mlsanity.com/paper/2202.09729/) (2022)
- [Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation](https://mlsanity.com/paper/1806.03185/) (2018)
- [TasNet: time-domain audio separation network for real-time, single-channel speech separation](https://mlsanity.com/paper/1711.00541/) (2017)
- [MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement](https://mlsanity.com/paper/2104.03538/) (2021)
- [NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling](https://mlsanity.com/paper/2104.02321/) (2021)
- [AudioSR: Versatile Audio Super-resolution at Scale](https://mlsanity.com/paper/2309.07314/) (2023)
- [Neural Vocoder is All You Need for Speech Super-resolution](https://mlsanity.com/paper/2203.14941/) (2022)
- [NU-Wave 2: A General Neural Audio Upsampling Model for Various Sampling Rates](https://mlsanity.com/paper/2206.08545/) (2022)
- [Upsampling artifacts in neural audio synthesis](https://mlsanity.com/paper/2010.14356/) (2020)
- [Audio Super Resolution using Neural Networks](https://mlsanity.com/paper/1708.00853/) (2017)
- [CU-Mamba: Selective State Space Models with Channel Learning for Image Restoration](https://mlsanity.com/paper/2404.11778/) (2024)
- [WSRGlow: A Glow-based Waveform Generative Model for Audio Super-Resolution](https://mlsanity.com/paper/2106.08507/) (2021)
- [NU-GAN: High resolution neural upsampling with GAN](https://mlsanity.com/paper/2010.11362/) (2020)

## Built on by (most cited first, 9 of 9)

- [Audio Super-Resolution with Latent Bridge Models](https://mlsanity.com/paper/2509.17609/) (2025)
- [UniverSR: Unified and Versatile Audio Super-Resolution via Vocoder-Free Flow Matching](https://mlsanity.com/paper/2510.00771/) (2025)
- [LongCat-AudioDiT: High-Fidelity Diffusion Text-to-Speech in the Waveform Latent Space](https://mlsanity.com/paper/2603.29339/) (2026)
- [Deep Active Speech Cancellation with Mamba-Masking Network](https://mlsanity.com/paper/2502.01185/) (2025)
- [SM3D: Mitigating Spectral Bias and Semantic Dilution in Point Cloud State Space Models](https://mlsanity.com/paper/2505.11099/) (2025)
- [Active Speech Enhancement: Active Speech Denoising Decliping and Deveraberation](https://mlsanity.com/paper/2505.16911/) (2025)
- [An overview of neural architectures for self-supervised audio representation learning from masked spectrograms](https://mlsanity.com/paper/2509.18691/) (2025)
- [ManchuTTS: Towards High-Quality Manchu Speech Synthesis via Flow Matching and Hierarchical Text Representation](https://mlsanity.com/paper/2512.22491/) (2025)
- [P2Flow: Phoneme-aware Progressive Flow Matching for Extreme Speech Super-Resolution](https://mlsanity.com/paper/2609.24138/) (2026)

## Abstract (arXiv)

Speech Super-Resolution (SSR) is a task of enhancing low-resolution speech signals by restoring missing high-frequency components. Conventional approaches typically reconstruct log-mel features, followed by a vocoder that generates high-resolution speech in the waveform domain. However, as mel features lack phase information, this can result in performance degradation during the reconstruction phase. Motivated by recent advances with Selective State Spaces Models (SSMs), we propose a method, referred to as Wave-U-Mamba that directly performs SSR in time domain. In our comparative study, including models such as WSRGlow, NU-Wave 2, and AudioSR, Wave-U-Mamba demonstrates superior performance, achieving the lowest Log-Spectral Distance (LSD) across various low-resolution sampling rates, ranging from 8 to 24 kHz. Additionally, subjective human evaluations, scored using Mean Opinion Score (MOS) reveal that our method produces SSR with natural and human-like quality. Furthermore, Wave-U-Mamba achieves these results while generating high-resolution speech over nine times faster than baseline models on a single A100 GPU, with parameter sizes less than 2% of those in the baseline models.