Transcription of automated phone message Speech to Text

I need to be able to invoke a phone call to a number (example +442890935124) and transcribe the first 2 minutes automated message that is played. This output is then to be stored in a file for further processing / parsing. The invocation will generally be automated with the required phoneNumber being passed.
We are looking at using the azure stack and the following resources (but are generally needing code that will work):
Logic apps - send the phone number
Azure functions, make call and handle call events
Azure Communication Services and call automation
Azure cognitive services and speech service
Azure event grid
Blob storage

We are open to a different implementation methods (app service or similar) but it has to be done in Azure (Customer requirement).
This is a backend app and there is no need for any UI type of work.

We need a fully tested, commented and documented app / set of files that can be deployed to my customers azure stack.