-
Notifications
You must be signed in to change notification settings - Fork 30
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Using RecorderService manim-voiceover hangs (before or after releasing the 'r' key?) #56
Comments
I can't reproduce this locally. Can you maybe insert some breakpoints ( |
Well, doing like you asked I determined that manim-voiceover hangs in services/recorder/utility.py between line 160 (reached) and line 163 (never reached). It reaches line 179, therefore it seems that MyListener() is not able to recognise pression of 'r' key. i.e. that self.listener.key_pressed is always false! |
Doing more debugging I determined that MyListener.on_press() is never called (the listener gets initialised and started, though), therefore there is no chance for key_pressed to become True, and its value remains None! |
Thank you, I’ll investigate this sooon |
Ok, I found the origin of the issue: the fact is that pynput is not meant to work on Wayland but only on Xorg. When I switched to Ubuntu on Xorg all worked as expected. Knowing this, you should either search for an alternative to pynput that works also on Wayland (such as this which, unfortunately, is currently unmaintained), or warn users to avoid Wayland when using manim-voiceover on Ubuntu... |
Then this might relate to #44, you can follow the discussion there. (Note that I had issues with Gradio, and that's why I didn't merge until now) The CLI based approach was a quick hack to get the MVP going, and the ideal solution would be a standalone UI that works without cross platform compatilibility issues. I found the following options: https://github.com/PySimpleGUI/PySimpleGUI We could also go for an Electron app or locally hosted web app like Jupyter Notebook but then it would be more complicated to ship Python and JS code in the same package, albeit more future-proof. Open to suggestions. |
Description of bug / unexpected behavior
I render a file with CoquiService and both manim and manim-voicover work correctly. Then I pass to RecorderService, I select the input device (e.g. 13 - default) and start recording pressing the 'r' key. When I release the 'r' key manim-voiceover hangs as if lost in an infinite idle cycle. In fact, inspecting folder ./media/voiceovers I see that no file has been produced, therefore I suspect that manim-voiceover hangs waiting the user to press 'r'...
Expected behavior
Manim-voiceover should start recording the voiceover as soon as the user presses 'r'. After releasing the 'r' key manim-voiceover should ask to choose from the following options:
l to [l]isten to the recording
r to [r]e-record
a to [a]ccept the recording
How to reproduce the issue
Code for reproducing the problem
Additional media files
Images/GIFs
Logs
Terminal output
System specifications
System Details
python/py/python3 --version
): 3.10.11pip list
):LaTeX details
FFMPEG
Output of
ffmpeg -version
:Additional comments
The text was updated successfully, but these errors were encountered: