Does Custom Neural Voice support Viseme output?

Lingga Pradipta 0

I could not find any documentation on the SSML supported by custom neural voice. I would like to know if Viseme events are supported.

kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-08T15:38:22.19+00:00

Hi Lingga Pradipta

Greetings & Welcome to the Microsoft Q&A forum! Thank you for sharing your query.

Currently, the viseme events feature is exclusively available for the en-US English (United States) neural voices. This means that if you are considering utilizing your own custom voice or audio, unfortunately, this feature will not be accessible to you at this time. As of now, there are no updates regarding the extension of viseme event support to other custom neural voices. However, should there be any developments or enhancements in this area, we will ensure to announce them through the Azure update pages or relevant blogs. can you please kindly refer this.

I hope this helps you. Thank you.
Lingga Pradipta 0 Reputation points

2025-01-09T12:25:56.4433333+00:00

I'm getting mixed messages since some answers say that you can't. But some like this imply that it's possible.
kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-15T08:04:06.4466667+00:00
Hi Lingga Pradipta

sorry for delay

Viseme events are supported when using SSML with custom neural voices. To utilize this feature, you need to include the mstts:viseme element in your SSML document. This allows the system to generate visual representations of phonemes as they are spoken, which is particularly useful for applications involving animation or lip-syncing can you please refer this

To implement viseme events in your SSML, follow these steps:

This element must be added to your SSML to request viseme data.

Use the Viseme Received event in the Speech SDK to capture viseme data during speech synthesis.

Example of Using Viseme in SSML

Here’s a basic example of how you might structure your SSML to include viseme events:

xml

<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="en-US">

<voice name="my-custom-voice"> <mstts:viseme>...</mstts:viseme> This is the text that is spoken. </voice>

</speak>

I hope this helps you Thank you.
kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-16T09:32:25.79+00:00

Hi Lingga Pradipta

Following up to see if the above response was helpful.
kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-17T05:29:11.16+00:00

Hi Lingga Pradipta

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.
Lingga Pradipta 0 Reputation points

2025-01-17T10:29:42.3766667+00:00

I think your solution will work.
kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-20T04:34:57.4133333+00:00

Hi Lingga Pradipta

Following up to see if the response was helpful.
kothapally Snigdha 950 Reputation points Microsoft Vendor

2025-01-21T05:34:34.7966667+00:00

Hi Lingga Pradipta

We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution, please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.

Share via

Does Custom Neural Voice support Viseme output?

Your answer