You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/cognitive-services/Speech-Service/speech-synthesis-markup.md
+10-9Lines changed: 10 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -133,6 +133,8 @@ Styles, style degree, and roles are supported for a subset of neural voices. If
133
133
134
134
You use the `mstts:express-as` element to express emotions like cheerfulness, empathy, and calm. You can also optimize the voice for different scenarios like customer service, newscast, and voice assistant.
135
135
136
+
For a list of supported styles per neural voice, see [supported voice styles and roles](language-support.md#voice-styles-and-roles).
137
+
136
138
**Syntax**
137
139
138
140
```xml
@@ -158,7 +160,7 @@ The following table has descriptions of each supported style.
158
160
159
161
|Style|Description|
160
162
|-----------|-------------|
161
-
|`style="advertisement-upbeat"`|Expresses an excited and high-energy tone for promoting a product or service.|
163
+
|`style="advertisement_upbeat"`|Expresses an excited and high-energy tone for promoting a product or service.|
162
164
|`style="affectionate"`|Expresses a warm and affectionate tone, with higher pitch and vocal energy. The speaker is in a state of attracting the attention of the listener. The personality of the speaker is often endearing in nature.|
163
165
|`style="angry"`|Expresses an angry and annoyed tone.|
164
166
|`style="assistant"`|Expresses a warm and relaxed tone for digital assistants.|
@@ -187,15 +189,17 @@ The following table has descriptions of each supported style.
187
189
|`style="sad"`|Expresses a sorrowful tone.|
188
190
|`style="serious"`|Expresses a strict and commanding tone. Speaker often sounds stiffer and much less relaxed with firm cadence.|
189
191
|`style="shouting"`|Speaks like from a far distant or outside and to make self be clearly heard|
190
-
|`style="sports-commentary"`|Expresses a relaxed and interesting tone for broadcasting a sports event.|
191
-
|`style="sports-commentary-excited"`|Expresses an intensive and energetic tone for broadcasting exciting moments in a sports event.|
192
+
|`style="sports_commentary"`|Expresses a relaxed and interesting tone for broadcasting a sports event.|
193
+
|`style="sports_commentary_excited"`|Expresses an intensive and energetic tone for broadcasting exciting moments in a sports event.|
192
194
|`style="whispering"`|Speaks very softly and make a quiet and gentle sound|
193
195
|`style="terrified"`|Expresses a very scared tone, with faster pace and a shakier voice. It sounds like the speaker is in an unsteady and frantic status.|
194
196
|`style="unfriendly"`|Expresses a cold and indifferent tone.|
195
197
196
198
### Style degree
197
199
198
-
The intensity of speaking style can be adjusted to better fit your use case. You specify a stronger or softer style with the `styledegree` attribute to make the speech more expressive or subdued. Speaking style degree adjustments are supported for Chinese (Mandarin, Simplified) neural voices.
200
+
The intensity of speaking style can be adjusted to better fit your use case. You specify a stronger or softer style with the `styledegree` attribute to make the speech more expressive or subdued.
201
+
202
+
For a list of neural voices that support speaking style degree, see [supported voice styles and roles](language-support.md#voice-styles-and-roles).
199
203
200
204
**Syntax**
201
205
@@ -220,12 +224,9 @@ This SSML snippet illustrates how the `styledegree` attribute is used to change
220
224
221
225
### Role
222
226
223
-
Apart from adjusting the speaking styles and style degree, you can also adjust the `role` parameter so that the voice imitates a different age and gender. For example, a male voice can raise the pitch and change the intonation to imitate a female voice, but the voice name won't be changed. Role adjustments are supported for these Chinese (Mandarin, Simplified) neural voices:
227
+
Apart from adjusting the speaking styles and style degree, you can also adjust the `role` parameter so that the voice imitates a different age and gender. For example, a male voice can raise the pitch and change the intonation to imitate a female voice, but the voice name won't be changed.
224
228
225
-
*`zh-CN-XiaomoNeural`
226
-
*`zh-CN-XiaoxuanNeural`
227
-
*`zh-CN-YunxiNeural`
228
-
*`zh-CN-YunyeNeural`
229
+
For a list of supported roles per neural voice, see [supported voice styles and roles](language-support.md#voice-styles-and-roles).
0 commit comments