Audio Super-Resolution
14 papers with code • 4 benchmarks • 3 datasets
AUDIO SUPER-RESOLUTION or speech bandwidth extension (Upsampling Ratio = 2)
Benchmarks
These leaderboards are used to track progress in Audio Super-Resolution
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
|||||
|
|||||
|
VCTK
MedleyDB 2.0
Latest papers
AudioSR: Versatile Audio Super-resolution at Scale
haoheliu/versatile_audio_super_resolution
•
•
Audio super-resolution is a fundamental task that predicts high-frequency components for low-resolution audio, enhancing audio quality in digital applications.
AERO: Audio Super Resolution in the Spectral Domain
slp-rl/aero
•
•
We optimize the model using both time and frequency domain loss functions.
Nonparallel High-Quality Audio Super Resolution with Domain Adaptation and Resampling CycleGANs
Neural audio super-resolution models are typically trained on low- and high-resolution audio signal pairs.
CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement
Convolution-augmented transformers (Conformers) are recently proposed in various speech-domain applications, such as automatic speech recognition (ASR) and speech separation, as they can capture both local and global dependencies.
NU-Wave 2: A General Neural Audio Upsampling Model for Various Sampling Rates
maum-ai/nuwave
•
•
Conventionally, audio super-resolution models fixed the initial and the target sampling rates, which necessitate the model to be trained for each pair of sampling rates.
Neural Vocoder is All You Need for Speech Super-resolution
In this paper, we propose a neural vocoder based speech super-resolution method (NVSR) that can handle a variety of input resolution and upsampling ratios.
Learning Continuous Representation of Audio for Arbitrary Scale Super Resolution
ml-postech/lisa
•
•
To obtain a continuous representation of audio and enable super resolution for arbitrary scale factor, we propose a method of implicit neural representation, coined Local Implicit representation for Super resolution of Arbitrary scale (LISA).
TUNet: A Block-online Bandwidth Extension Model based on Transformers and Self-supervised Pretraining
nxtproduct/tunet
•
•
We introduce a block-online variant of the temporal feature-wise linear modulation (TFiLM) model to achieve bandwidth extension.
Self-Attention for Audio Super-Resolution
ncarraz/AFILM
•
•
Convolutions operate only locally, thus failing to model global interactions.
NU-Wave: A Diffusion Probabilistic Model for Neural Audio Upsampling
maum-ai/nuwave
•
•
In this work, we introduce NU-Wave, the first neural audio upsampling model to produce waveforms of sampling rate 48kHz from coarse 16kHz or 24kHz inputs, while prior works could generate only up to 16kHz.
