Audio to Video Generation Tool

Final deployed Audio to Video webapp link : Techvistaraconvertor

With the help of Replit AI and deployed to Webapp

• As there is currently a lot of audio content being released, such as Overview by NotebookLM from Google recently, building a tool to convert voice to a video format for uploading to certain video-only platforms is valuable.
• Naming this tool as : Techvistaraconvertor.
• The link to try out the tool is provided at the top: Techvistaraconvertor
• This exercise demonstrates that constant iteration and developing a solid framework for cloud deployment will give rise to many tools in the coming days of AI.
• https://blog.google/technology/ai/notebooklm-audio-overviews/

Initial Look of the Tool in Replit AI

• Using an audio file, in this case, we are using Google's NotebookLM downloaded audio.
• A custom image to appear on the video while playing.
• Generating the video with resolution selection.

Replit Development interface

Working version

• Enhanced the convertor to list the already available audio and images
• The instructions in setting up created by replit AI : README.md

Creating and linking Azure webapp to deploy the app

Code repo : https://github.com/dinesh-coderepo/techvistaraconvertor

• Linked the web app to the repo to enable CI/CD for incremental changes
• More details in the above code repo regarding the workflow deployment script: Link
• Deployed to the webapp: https://audiotovideo.azurewebsites.net/

• The compute selected for the app service is very minimal hence we can expect resource crunch while video generation, but this exercise demonstrates building a tool for the need of a use case.