How much compute hours is required to fine tune approximately 5-10 hours of audio data on custom speech to text fine tune

I'd like to know if there is any relevancy or blogs available to get to know the hours of audio that'd be processed based on custom speech training hours.

For e.g.
10 hours of audio equivalent to 2-3 hours of compute hours. anything available to get the information for the following?

Thanks and regards!!!

https://azure.microsoft.com/en-in/pricing/calculator/