You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository was archived by the owner on Feb 7, 2024. It is now read-only.
Repository navigation
This repository was archived by the owner on Feb 7, 2024. It is now read-only.
The *.wav files I am trying to read are quite large (about 100e6 samples) and I get a memory error when trying to use the gain function, and the function Signal.from_wav takes really long to execute.
Any tip to be able to read the files per chunks or to read them faster?
PySoundFile has a way to read per slices the wav files, and their reading algorithm is a bit faster, but then it takes again so long to convert it to an acoustics.Signal object.
I could also maybe try to implement it myself if there is no code prepared for it or any trick I am missing.
Any help will be really appreciated! :)
I suppose the Signal class could be improved by supporting numpy.memmap. Anyway, the way it is now, the Signal class is not intended for long signals requiring in-core operations.
The *.wav files I am trying to read are quite large (about 100e6 samples) and I get a memory error when trying to use the gain function, and the function Signal.from_wav takes really long to execute.
Any tip to be able to read the files per chunks or to read them faster?
PySoundFile has a way to read per slices the wav files, and their reading algorithm is a bit faster, but then it takes again so long to convert it to an acoustics.Signal object.
I could also maybe try to implement it myself if there is no code prepared for it or any trick I am missing.
Any help will be really appreciated! :)