wv-render stands for wave render, although it actually renders a spectral representation of the audio that was submitted. It is not on-par scientific accuracy, and is for creative purposes only.
Idea for wv-render originally begun out as a C program that used Cairo to render submitted audio into a waveform, but it wasn't extended further and left in dark.
This project ressurects the aforementioned idea, and instead of using C directly, uses Python along with various high performance libraries. Native Python code is kept to a minimum, and most of the mathematical work is done by other libraries.
Only video is exported, audio isn't, so to recombine the audio, you could use FFmpeg.