Hi! This is the support thread for the Multimodal Virtual Assistant project. Any major updates we push to the project will be announced here. Further, feel free to discuss, raise issues, and ask for assistance in this thread.
Please keep discussion in this thread project-related. Any issues with the Workbench application should be raised as a standalone thread. Thanks!
Project is live! Note any bugs or issues you run into in this thread. Thanks!
Error when uploading any kind of youtube video. Any ideas ? I can actually see the file created in /project/data/video_data
warning
Video URL upload failed, try again later. Exception: MoviePy error: the file /project/data/video_data/input_vid_3.mp4 could not be found! Please check that you entered the correct path
I want to further customize this project. I’m interested in removing the access to OpenAI or Tavily APIs. We currently have Dell AI Factory with a Dell PowerEdge R760XA server. What kind of customization could I do?
any resolution on this as I am facing similar issue?
any resolution on this as I am facing similar issue?
Hi, thanks for reaching out. It sounds like there may be a filename collision during the upload which threw the db into a bad state.
You can try to delete the video_data directory and restart the project/project container to reset and retry the upload.
Those secrets are built into the default agent graph for the project so it may take some effort to remove them from the structure of the project. You can try taking a look at the graph.py file to get started, for example, as this defines the Tavily tool calling action.
4/17/2025
gradio version bump