issue with the Azure Text-to-Speech (TTS) service when using the SSML

IJazulhassan 0

Dear Microsoft Azure Cognitive Services Support Team,

I hope you are doing well.

I am letting you know about an issue with the Azure Text-to-Speech (TTS) service when using the SSML `

santoshkc 8,775 Reputation points Microsoft Vendor

2024-08-13T13:23:52.5133333+00:00

Hi @IJazulhassan,

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

The issue you're encountering with the whispering style in SSML might stem from differences in how the Azure Voice Generator tool and the Azure TTS API handle SSML tags. The API might not fully support applying the <mstts:express-as style="whispering"> tag to individual words within a sentence, even though it works in the Voice Generator tool.

To work around the issue, try applying the whispering style to larger chunks of text rather than just individual words. For instance, group the word you want to emphasize with its surrounding context into a larger segment and apply the style to that entire segment. This approach ensures that the style is applied consistently across the text.

I hope you understand. Do let us know if you have any further queries.

Do click Accept Answer and Yes for was this answer helpful.
santoshkc 8,775 Reputation points Microsoft Vendor

2024-08-14T12:00:04.3566667+00:00

Hi @IJazulhassan,

Did you got any chance to check the below response?
santoshkc 8,775 Reputation points Microsoft Vendor

2024-08-16T04:44:28.92+00:00

Hi @IJazulhassan,

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.

Do click Accept Answer and Yes for the below was answer helpful.
IJazulhassan 0 Reputation points

2024-09-24T07:11:27.6166667+00:00

Hi @santoshkc ,

Thank you for the response and pardon me for the late reply, kindly can you please share any documentation or rules related to such scenarios like what are the cases in which the speech services (TTS) allow the use of single or pair of words for <mstts:express-as style=[style]> SSML tag. I tried different scenarios which sometimes randomly worked and sometimes didn't both individually and for many words.
Please let me know if anything in detail is required or to discuss more, looking forward to your reply.Thanks
santoshkc 8,775 Reputation points Microsoft Vendor

2024-09-24T12:55:08.72+00:00

Hi @IJazulhassan,

Thank you for your follow up query.

For Azure Text-to-Speech (TTS) using SSML, specific guidelines and limitations exist regarding how certain SSML tags, like mstts:express-as style="whispering", can be applied. The behaviour can vary depending on the tag's application to individual words or larger segments.

You can refer to SSML overview for Azure Speech Services, which covers best practices, supported tags, and limitations.

Typically, SSML tags like mstts:express-as are designed to be applied to entire phrases or sentences rather than single words. Applying the tag to single words might not consistently work due to how the TTS engine processes the voice style, which may explain why your usage has been unpredictable. The general recommendation is to apply these tags to larger segments to ensure consistent results.

I hope you understand. Thank you.

Share via

issue with the Azure Text-to-Speech (TTS) service when using the SSML

Your answer