Optional
roleAge and gender role for voice modification.
Simulates different speaker characteristics. Only supported by certain voices.
Emotional or speaking style to apply. (Required)
The available styles depend on the voice being used. Common categories include emotions (cheerful, sad, angry), professional styles (newscast, customerservice), and special effects (whispering, shouting).
Optional
styledegreeIntensity of the style expression.
Controls how strongly the style is applied. Range: "0.01" (minimal) to "2" (double intensity)
Configuration options for express-as elements (Azure-specific).
Controls emotional expression and speaking styles for neural voices that support these features. Allows for nuanced emotional delivery and role-playing scenarios.
Example
See
https://docs.microsoft.com/azure/cognitive-services/speech-service/speech-synthesis-markup#express-as-element Express-As Documentation