This project demonstrates how to transcribe audio from your media.
-
With this tool, you can then tidy up the transcribed text into bite-sized chunks and feed back into Azure Custom Voice service.
-
In a few simple steps you can recreate the voice and make the original speaker say anything you want!
-
You can take as little as ten minutes audio from friends, family or a video of your favourite star.
- This project shows how to use Azure Speech service, to transcribe speech to text.
- You can use many types of media, because the project will strip the Wave (sound) file out of a video like mp4 or mov, whatever NAudio is compatible with.
- You then curate the snippets and feed them back into Azure Custom Voice service, to generate a surprisingly similar sounding voice!
- Then, you can feed text back in, to generate your own audio files using the custom voice!
- You can also export the "timed text" in a couple of popular formats. The TTML format will work with the baked in Windows 10 "Films & TV" app.
All you need is a FREE Azure subscription and a FREE Speech Service key.
Read all about this project on TechNet Wiki
This Cognitive Services implementation example is a contribution to the Microsoft IT Implementer initiative. Demystifying the latest tech and demonstrating how to actually implement it yourself!
I am also very proud to have been awarded the MVP (Most Valuable Professional) award for three years.
But of course, just passing on fun stuff that I learn along the way, is the main driver and a great pleasure.
Pete Laker
Microsoft Azure MVP, Microsoft IT Implementer, TechNet Wiki Community Council