I am using Whisper (Speech Recognition model) to translate speeches. I have already configured and use whisper in Google Colab. Now I am developing a simple User interface (Web App) for better user experience. I have already deployed Whisper model on Baseten (https://www.baseten.co/) and now could call the model via an API to get back the response but Whisper model take time to get back the response depending upon the file size to process but usually web hosting have maximum script execution time limit so how could we wait for the response to get back from the deployed whisper model before the termination of our backend web script written in (PHP,Python,Node js etc). Should we purchase a virtual server/dedicated server to host such AI based Web apps or is it possible in shared web hosting as well. What backend web technology stack would be suitable for such use cases? I am looking forward to hearing from any one who has expertise in building AI based Web apps.
I tried free hosting provided by Infinity free but it has limitations. So I want opinion from the community before going to a paid hosting.