Customize text to speech avatar gestures with SSML (preview)
Note
Text to speech avatar is currently in public preview. This preview is provided without a service-level agreement, and is not recommended for production workloads. Certain features might not be supported or might have constrained capabilities. For more information, see Supplemental Terms of Use for Microsoft Azure Previews.
The Speech Synthesis Markup Language (SSML) with input text determines the structure, content, and other characteristics of the text to speech output. Most SSML tags can also work in text to speech avatar. Furthermore, text to speech avatar batch mode provides avatar gestures insertion ability by using the SSML bookmark element with the format <bookmark mark='gesture.*'/>
.
A gesture starts at the insertion point in time. If the gesture takes more time than the audio, the gesture is cut at the point in time when the audio is finished.
Bookmark example
The following example shows how to insert a gesture in the text to speech avatar batch synthesis with SSML.
<speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="en-US">
<voice name="en-US-AvaMultilingualNeural">
Hello <bookmark mark='gesture.wave-left-1'/>, my name is Ava, nice to meet you!
</voice>
</speak>
In this example, the avatar will start waving their hand at the left after the word "Hello".
Supported prebuilt avatar characters, styles, and gestures
The full list of prebuilt avatar supported gestures provided here can also be found in the text to speech avatar portal.
Characters | Styles | Gestures |
---|---|---|
Harry | business | 123 calm-down come-on five-star-reviews good hello introduce invite thanks welcome |
Harry | casual | 123 come-on five-star-reviews gong-xi-fa-cai good happy-new-year hello please welcome |
Harry | youthful | 123 come-on down five-star good hello invite show-right-up-down welcome |
Jeff | business | 123 come-on five-star-reviews hands-up here meddle please2 show silence thanks |
Jeff | formal | 123 come-on five-star-reviews lift please silence thanks very-good |
Lisa | casual-sitting | numeric1-left-1 numeric2-left-1 numeric3-left-1 thumbsup-left-1 show-front-1 show-front-2 show-front-3 show-front-4 show-front-5 think-twice-1 show-front-6 show-front-7 show-front-8 show-front-9 |
Lisa | graceful-sitting | wave-left-1 wave-left-2 thumbsup-left show-left-1 show-left-2 show-left-3 show-left-4 show-left-5 show-right-1 show-right-2 show-right-3 show-right-4 show-right-5 |
Lisa | graceful-standing | |
Lisa | technical-sitting | wave-left-1 wave-left-2 show-left-1 show-left-2 point-left-1 point-left-2 point-left-3 point-left-4 point-left-5 point-left-6 show-right-1 show-right-2 show-right-3 point-right-1 point-right-2 point-right-3 point-right-4 point-right-5 point-right-6 |
Lisa | technical-standing | |
Lori | casual | 123-left a-little beg calm-down come-on five-star-reviews good hello open please thanks |
Lori | graceful | 123-left applaud come-on introduce nod please show-left show-right thanks welcome |
Lori | formal | 123 come-on come-on-left down five-star good hands-triangle hands-up hi hopeful thanks |
Max | business | a-little-bit click-the-link display-number encourage-1 encourage-2 five-star-praise front-right good-01 good-02 introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 left lower-left number-one press-both-hands-down-1 press-both-hands-down-2 push-forward raise-ones-hand right say-hi shrug-ones-shoulders slide-from-left-to-right slide-to-the-left thanks the-front top-middle-and-bottom-left top-middle-and-bottom-right upper-left upper-right welcome |
Max | casual | a-little-bit applaud click-the-link display-number encourage-1 encourage-2 five-star-praise front-left good-1 good-2 hello introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 introduction-to-products-4 left length nodding number-one press-both-hands-down raise-ones-hand right right-front shrug-ones-shoulders slide-from-left-to-right slide-to-the-left thanks the-front upper-left upper-right welcome |
Max | formal | a-little-bit click-the-link display-number encourage-1 encourage-2 five-star-praise front-left front-right good-1 good-2 introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 left lower-left lower-right press-both-hands-down push-forward right say-hi shrug-ones-shoulders slide-from-left-to-right slide-to-the-left the-front top-middle-and-bottom-right upper-left upper-right |
Meg | formal | a-little-bit click-the-link display-number encourage-1 encourage-2 five-star-praise front-left front-right good-1 good-2 hands-forward introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 left number-one press-both-hands-down-1 press-both-hands-down-2 right say-hi shrug-ones-shoulders slide-from-left-to-right the-front upper-left upper-right |
Meg | casual | a-little-bit click-the-link cross-hand display-number encourage-1 encourage-2 five-star-praise front-left front-right good-1 good-2 handclap introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 left length lower-left lower-right number-one press-both-hands-down right say-hi shrug-ones-shoulders slide-from-right-to-left slide-to-the-left spread-hands the-front top-middle-and-bottom-left top-middle-and-bottom-right upper-left upper-right |
Meg | business | a-little-bit encourage-1 encourage-2 five-star-praise front-left front-right good-1 good-2 introduction-to-products-1 introduction-to-products-2 introduction-to-products-3 left length number-one press-both-hands-down-1 press-both-hands-down-2 raise-ones-hand right say-hi shrug-ones-shoulders slide-from-left-to-right slide-to-the-left spread-hands thanks the-front upper-left |
Only the casual-sitting
style is supported via the real-time text to speech API. Gestures are only supported with the batch synthesis API and aren't supported via the real-time API.
Next steps
Feedback
https://aka.ms/ContentUserFeedback.
Coming soon: Throughout 2024 we will be phasing out GitHub Issues as the feedback mechanism for content and replacing it with a new feedback system. For more information see:Submit and view feedback for