search

madhavmk / Noise2Noise-audio_denoising_without_clean_training_data

Source code for the paper titled "Speech Denoising without Clean Training Data: a Noise2Noise Approach". Paper accepted at the INTERSPEECH 2021 conference. This paper tackles the problem of the heavy dependence of clean speech data required by deep learning based audio denoising methods by showing that it is possible to train deep speech denoising networks using only noisy speech samples.
MIT License
177 stars 42 forks source link

Suspicious requirements.txt file #6

Open alimaslax opened 1 year ago

alimaslax commented 1 year ago

What's with the requirements? olefile, webSockets and Certifi?

pysocks, pycparser, etc...

Security and Cryptography: The inclusion of packages like cryptography and argon2-cffi suggests that the program may involve cryptographic operations, such as encryption, decryption, hashing, or secure password handling.

madhavmk commented 1 year ago

You are right @alimaslax . There are some redundant requirements at the moment (this repo originally included a production webserver code to serve denoised audio). If you have a minimal conda env requirements list, please make a PR, and I can merge it 🙂