MICHAEL CHOURDAKIS
MICHAEL CHOURDAKIS
The idea is to display a ghost note on mouse over so when the user clicks, I create a note at that position. The problem that is when I actually...
This is interesting, I tried to build it and done it with some hassle (using import soundfile, some type conversions), then injected a new file into training_data\raw_audio\vocal_tracks as survive_4_vocals.wav, then...
It would be nice to also implement the erase functions.
I 'm calling hxcmod_fillbuffer in a loop all right, how do I know when to stop because there's nothing more? How do i start at a specific position?
It's very nice. Can it run without the call to std::make_shared to avoid a memory allocation?
It seems nice, but using OpenSSL locks me out (uneccessary overhead). It's a Windows project, so can it make it use Windows API instead of OpenSSL and ASIO? Good job.
Give us some example on raw float samples in/out, available parameters to modify etc, I want to create a VST from this one.
This is fine. I'm able to restore speech. Now a non-speech recording fixer could be really useful here. Any hints on how to train my own one on that?
It is excellent, I use it in my video sequencer to upscale videos. Would it be possible to have the executable able to be fed images from std in and...
``` import numpy as np import soundfile as sf from tqdm import tqdm from nara_wpe.wpe import wpe from nara_wpe.wpe import get_power from nara_wpe.utils import stft, istft, get_stft_center_frequencies from nara_wpe import...