Is there access to additional properties from the Speech Studio Voice Gallery via SDK or API?

Question

Is there access to additional properties from the Speech Studio Voice Gallery via SDK or API?

Jones, Andrew 0

The Speech Studio Voice Gallery includes additional information such as age groups, personality, and scenarios. However, when using the SDK or REST API to retrieve voices, only a limited set of properties is available. Are these additional properties accessible through the SDK or API? The /texttospeech/acc/v3.0-beta1/vcg/voices API endpoint used by the gallery is undocumented, and attempts to call it with a subscription ID result in a 404 error.

santoshkc 15,355 Reputation points Microsoft External Staff Moderator

2025-02-10T11:08:02.31+00:00

Hi @Jones, Andrew,

Following up to see if the given response answers your query. Thank you.
Jones, Andrew 0 Reputation points

2025-02-10T13:55:15.2966667+00:00

@santoshkc Yes - you've clarified that this additional information is unavailable. I am aware of the mstts:express-as tag but the documentation seems to indicate that not all voices support all styles and roles. The speech gallery includes the list of supported (or possibly recommended) roles/styles for each voice which would be useful information.

Thanks for following up!
santoshkc 15,355 Reputation points Microsoft External Staff Moderator

2025-02-12T13:16:51.6033333+00:00

Hi @Jones, Andrew,

The additional voice properties (age, personality, scenarios, styles) from Speech Studio Voice Gallery are not available via API/SDK. You can manually compile this info or test mstts:express-as values in SSML to check supported styles.

Thank you.

1 answer

Your answer

santoshkc 15,355 Reputation points Microsoft External Staff Moderator

2025-02-10T11:08:02.31+00:00

Hi @Jones, Andrew,

Following up to see if the given response answers your query. Thank you.
Jones, Andrew 0 Reputation points

2025-02-10T13:55:15.2966667+00:00

@santoshkc Yes - you've clarified that this additional information is unavailable. I am aware of the mstts:express-as tag but the documentation seems to indicate that not all voices support all styles and roles. The speech gallery includes the list of supported (or possibly recommended) roles/styles for each voice which would be useful information.

Thanks for following up!
santoshkc 15,355 Reputation points Microsoft External Staff Moderator

2025-02-12T13:16:51.6033333+00:00

Hi @Jones, Andrew,

The additional voice properties (age, personality, scenarios, styles) from Speech Studio Voice Gallery are not available via API/SDK. You can manually compile this info or test mstts:express-as values in SSML to check supported styles.

Thank you.

Answer 1

Hi @Jones, Andrew,

Thank you for reaching out to Microsoft Q&A forum!

Currently, the Azure Speech SDK and REST API only provide basic properties for voices, such as name, locale, and gender. The additional details shown in the Speech Studio Voice Gallery are not exposed through the public API or SDK.

The /texttospeech/acc/v3.0-beta1/vcg/voices endpoint you mentioned is likely an internal, undocumented API, which is why it's not accessible.

If you’re looking for specific voice characteristics like styles or roles, you can use SSML with the mstts:express-as tag. This lets you adjust the speaking style for supported neural voices.

Thank you.

Share via

Is there access to additional properties from the Speech Studio Voice Gallery via SDK or API?

1 answer

Your answer