Merge pull request #271 from MicrosoftDocs/main

Publish to live, Friday 4 AM PST, 9/13

Author: ttorble

articles/ai-services/speech-service/openai-voices.md

@@ -36,7 +36,7 @@ You can choose from the following text to speech voices in Azure AI services:
 
 ## OpenAI text to speech voices via Azure OpenAI Service or via Azure AI Speech?
 
-If you want to use OpenAI text to speech voices, you can choose whether to use them via [Azure OpenAI](../openai/text-to-speech-quickstart.md) or via [Azure AI Speech](./get-started-text-to-speech.md#openai-text-to-speech-voices-in-azure-ai-speech). In either case, the speech synthesis result is the same. To explore OpenAI text to speech voices through Azure AI Speech, visit the [Voice Gallery](https://speech.microsoft.com/portal/voicegallery).
+If you want to use OpenAI text to speech voices, you can choose whether to use them via [Azure OpenAI](../openai/text-to-speech-quickstart.md) or via [Azure AI Speech](./get-started-text-to-speech.md#openai-text-to-speech-voices-in-azure-ai-speech). You can visit the [Voice Gallery](https://speech.microsoft.com/portal/voicegallery) to listen to samples of Azure OpenAI voices or synthesize speech with your own text using the [Audio Content Creation](https://speech.microsoft.com/portal/audiocontentcreation). The audio output is identical in both cases, with only a few feature differences between the two services. See the table below for details.
 
 Here's a comparison of features between OpenAI text to speech voices in Azure OpenAI Service and OpenAI text to speech voices in Azure AI Speech. 
 

articles/ai-services/speech-service/speech-container-ntts.md

@@ -184,7 +184,7 @@ For more information about `docker run` with Speech containers, see [Install and
 
 ### SSML voice element
 
-When you construct a neural text to speech HTTP POST, the [SSML](speech-synthesis-markup.md) message requires a `voice` element with a `name` attribute. The [locale of the voice](language-support.md?tabs=tts) must correspond to the locale of the container model. 
+When you construct a neural text to speech HTTP POST, the [SSML](speech-synthesis-markup.md) message requires a `voice` element with a `name` attribute. The [locale of the voice](language-support.md?tabs=tts) must correspond to the locale of the container model. The [SSML tag](speech-synthesis-markup-structure.md) support is consistent for [each text to speech voice](https://mcr.microsoft.com/product/azure-cognitive-services/speechservices/neural-text-to-speech/tags) both in the Azure cloud and the container environment.
 
 For example, a model that was downloaded via the `latest` tag (defaults to "en-US") would have a voice name of `en-US-AriaNeural`.