maxime-leiber / dstft
Differentiable short-time Fourier transform (DSTFT): Gradient-based parameters tuning for adaptive time-frequency representation. DSTFT is a neural network layer whose weights are its parameters (e.g. window and hop lengths).
☆33Updated 8 months ago
Alternatives and similar repositories for dstft:
Users that are interested in dstft are comparing it to the libraries listed below
- ☆27Updated 3 years ago
- Official implementation of Efficient Speech Separation Framework Based on Neural State-Space Models☆18Updated last year
- Repository of published DNN speech separation recipes for a number of datasets☆10Updated last year
- An 1D optimal transport inspired loss function in the spectral domain. Can be used for improving frequency localization/estimation in dif…☆19Updated 9 months ago
- ASiT: Audio Spectrogram vIsion Transformer for General Audio Representation☆22Updated 10 months ago
- (TASLP 2022) Unsupervised speech enhancement using DVAEs☆21Updated last month
- Official Pytorch implementation of PULSE: Positive–Unlabelled Learning for audio Signal Enhancement (Best Paper Award at ICASSP 2023)☆41Updated last year
- ☆14Updated last year
- A neural speech codec based on discrete WavLM representations☆22Updated 5 months ago
- Speech enhancement in noisy and reverberant environments using deep neural networks☆20Updated 3 months ago
- [SLT'24] The official implementation of SSAMBA: Self-Supervised Audio Representation Learning with Mamba State Space Model☆108Updated 3 months ago
- An unofficial code reproduction of Channel Attention Dense U-Net for Multichannel Speech Enhancement☆11Updated last year
- Repo for source code of EBEN: Extreme Bandwidth Extension Network☆72Updated 2 weeks ago
- ☆10Updated 2 years ago
- SRTNet☆24Updated last year
- ☆64Updated 3 months ago
- Noise-Aware Speech Separation with Contrastive Learning☆16Updated 9 months ago
- context-aware Unet based on transformer for speech denoising☆23Updated 3 years ago
- A self-supervised speech denoising strategy named Only-Noisy Training (ONT), which solves the speech denoising problem with only noisy au…☆65Updated last year
- Transformer based Self-Attention for Complex Numbers☆12Updated 3 years ago
- ☆12Updated 2 years ago
- ☆17Updated 3 years ago
- TODO☆37Updated last year
- A 1D implementation of a deformable convolutional layer in PyTorch with a few tricks.☆40Updated last year
- ☆28Updated 8 months ago
- Official Implementation of "Inference and Denoise: Causal Inference-based Neural Speech Enhancement"☆27Updated last year
- A Diffusion Probabilistic Model for Target Sound Extraction☆36Updated 4 months ago
- DOSE: Diffusion Dropout with Adaptive Prior for Speech Enhancement, Conference on Neural Information Processing Systems (NeurIPS), 2023☆44Updated 2 weeks ago
- Generative Fixed-Filter Active Noise Control with CNN-Kalman Filtering☆24Updated 4 months ago
- DNN assisted Kalman filter for time domain speech enhancement☆21Updated 4 years ago