Signing up for Azure Cognitive Services Voice

To use Azure for text-to-speech (TTS), you will need to register for an Azure account and set up a subscription and cognitive services. Create a subscription here: https://azure.microsoft.com/free/

TO CREATE THE COGNITIVE SERVICES RESOURCE (AFTER CREATING A SUBSCRIPTION):

  1. Go to: https://ms.portal.azure.com/#create/Microsoft.CognitiveServicesAllInOne
  2. Enter a name for the resource. “Vera Sonos TTS” for example.
  3. Select the subscription (you’ll probably only have one to choose from)
  4. Location: choose the one that appears geographically nearest to you.
  5. Pricing tier, choose “S0” or whatever is available/cheapest.
  6. Resource group: choose an existing or create a new one. Doesn’t much matter what it is if this is all you’ll be using Azure for.
  7. Check the “OK/agree” box and then click “Create”.
  8. Wait for resource to be created. When it’s finished, click “Go to resource”.
  9. Enter the key shown into the Azure TTS config on the Sonos plugin.
  10. Look at the endpoint URL; it should be <region>.api.cognitive.microsoft.com … choose the region in the Azure TTS config for the Sonos plugin that matches the region shown.
2 Likes

I created an Azure account and went through all of these steps but the endpoint URL was different than what you have listed. It was https://verasonostts1.cognitiveservices.azure.com/… where verasonostts1 was the name for the resource. Can the endpoint URL be updated and if so how/where?

Install the hotfix version in the pinned post for the category. It will let you put in the endpoint, although I think I’ve since determined that you really don’t need it, they still seem to allow a default endpoint of <your-region>.api.cognitive.microsoft.com to work. Have you tried to use it as it is?

Sorry… updated the hotfix and it worked with no need to update the endpoint.

Awesome! I love it when a plan comes together! :slight_smile: