Merge pull request #249782 from sally-baolian/patch-147

Update record-custom-voice-samples.md

Author: prmerger-automator[bot]

articles/ai-services/speech-service/how-to-custom-voice-create-voice.md

@@ -86,7 +86,7 @@ To create a custom neural voice in Speech Studio, follow these steps for one of
 1. Select **Next**.
 1. Optionally, you can add additional custom speaking styles. The maximum number of custom styles varies by languages: `English (United States)` allows up to 10 custom styles, `Chinese (Mandarin, Simplified)` allows up to 4 custom styles, and `Japanese (Japan)` allows up to 5 custom styles.
     1. Select **Add a custom style** and thoughtfully enter a custom style name of your choice. This name will be used by your application within the `style` element of [Speech Synthesis Markup Language (SSML)](speech-synthesis-markup-voice.md#speaking-styles-and-roles). You can also use the custom style name as SSML via the [Audio Content Creation](how-to-audio-content-creation.md) tool in [Speech Studio](https://speech.microsoft.com/portal/audiocontentcreation).
-    1. Select style samples as training data. The style samples should be all from the same voice talent profile.
+    1. Select style samples as training data. Ensure that the training data for custom speaking styles comes from the same speaker as the data used to create the default style.
 1. Select **Next**.
 1. Select a speaker file with the voice talent statement that corresponds to the speaker in your training data.
 1. Select **Next**.

articles/ai-services/speech-service/record-custom-voice-samples.md

@@ -317,6 +317,8 @@ Create a reference recording, or *match file,* of a typical utterance at the beg
 
 The match file is especially important when you resume recording after a break or on another day. Play it a few times for the talent and have them repeat it each time until they're matching well.
 
+To record a corpus with a specific style, carefully choose scripts that showcase the desired style. During recording, ensure the voice talent maintains consistent in volume, tempo, pitch, and tone to achieve recordings that embody the intended style.
+
 Coach your talent to take a deep breath and pause for a moment before each utterance. Record a couple of seconds of silence between utterances. Words should be pronounced the same way each time they appear, considering context. For example, "record" as a verb is pronounced differently from "record" as a noun.
 
 Record approximately five seconds of silence before the first recording to capture the "room tone". This practice helps Speech Studio compensate for noise in the recordings.