Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Deep learning for audio denoising
| Date | Stars |
|---|---|
| 2026-07-31 | 765 |
| 2026-08-02 | 765 |
| 2026-08-06 | 766 |
Today
+1 stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# Speech-enhancement --- [](https://travis-ci.com/vbelz/Speech-enhancement) >Vincent Belz : [email protected] > >Published in towards data science : [Speech-enhancement with Deep learning](https://towardsdatascience.com/speech-enhancement-with-deep-learning-36a1991d3d8d) > ## Introduction **This project aims at building a speech enhancement system to attenuate environmental noise.** <img src="img/denoise_10classes.gif" alt="Spectrogram denoising" title="Speech enhancement"/> Audios have many different ways to be represented, going from raw time series to time-frequency decompositions. The choice of the representation is crucial for the performance of your system. Among time-frequency decompositions, Spectrograms have been proved to be a useful representation for audio processing. They consist in 2D images representing sequences of Short Time Fourier Transform (STFT) with time and frequency as axes, and brightness representing the strength of a frequency component at each time frame. In such they appear a natural domain to apply the CNNS architectures for images directly to sound. Between magnitude and phase spectrograms, magnitude spectrograms contain most the structure of the signal. Phase spectrograms appear to show only little temporal and spectral regularities. In this project, I will use magnitude spectrograms as a representation of sound (cf image below) in order to predict the noise model to be subtracted to a noisy voice spectrogram. <img src="img/sound_to_spectrogram.png" alt="sound representation" title="sound representation" /> The project is decomposed in three modes: `data creation`, `training` and `prediction`. ## Prepare the data To create the datasets for training, I gathered english speech clean voices and environmental noises from different sources. The clean voices were mainly gathered from [LibriSpeech](http://www.openslr.org/12/): an ASR corpus based on public domain audio books. I used as well some datas from [SiSec](https://sisec.inria.fr/sisec-2015/2015-two-channel-mixtures-of-speech-and-real-world-background-noise/). The environmental noises were gathered from [ESC-50 dataset](https://github.com/karoldvl/ESC-50) or [https://www.ee.columbia.edu/~dpwe/sounds/](https://www.ee.columbia.edu/~dpwe/sounds/). For this project, I focused on 10 classes of environmental noise: **tic clock**, **foot steps**, **bells**, **handsaw**, **alarm**, **fireworks**, **insects**, **brushing teeth**, **vaccum cleaner** and **snoring**. These classes are illustrated in the image below (I created this image using pictures from [https://unsplash.com](https://unsplash.com)). <img src="img/classes_noise.png" alt="classes of environmental noise used" title="classes of environmental noise" /> To create the datasets for training/validation/testing, audios were sampled at 8kHz and I extracted windows slighly above 1 second. I performed some data augmentation for the environmental noises (taking the windows at different times creates different noise windows). Noises have been blended to clean voices with a randomization of the noise level (between 20% and 80%). At the end, training data consisted of 10h of noisy voice & clean voice, and validation data of 1h of sound. To prepare the data, I recommend to create data/Train and data/Test folders in a location separate from your code folder. Then create the following structure as in the image below: <img src="img/structure_folder.png" alt="data folder structure" title="data folder structure" /> You would modify the `noise_dir`, `voice_dir`, `path_save_spectrogram`, `path_save_time_serie`, and `path_save_sound` paths name accordingly into the `args.py` file that takes the default parameters for the program. Place your noise audio files into `noise_dir` directory and your clean voice files into `voice_dir`. Specify how many frames you want to create as `nb_samples` in `args.py` (or
Excerpt of 13,898 characters
Read on GitHub
27
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:0ce237b42f9c1be6, topic:deep-learning