skip to Main Content

I am trying to reduce the gap between two sentence groups in Text to Speech in Azure console. The gap between sentences can be reduced by specifying sentenceboundarysilence i.e. sentenceboundarysilence="5ms", but I don’t know how to customize silence between sentence groups(as shown in below image). Basically, I want to specify the duration between the first sentence group (ending with "best game ever") and the second sentence group (starting with "Reason number one")

Azure

2

Answers


  1. Digging through SSML docs the main ways to customize pauses is with SentenaceBoundary, break, and silence. A break can be added anywhere in the text. Silence can only be used at the end of text. If you want to increase the pause between two groups of text, add a break at the end of the last sentence in a group of sentences. For Azure speech studio you can add a break by putting a time within square brackets like this:

    [600ms]
    

    Here are some docs on adding breaks/silence to SSML: https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/speech-synthesis-markup-structure#add-silence

    I don’t believe there is an option to reduce the pause between two sentence groups without creating some really messy SSML (you could set the time for sentence boundary to your minimum pause time, then add breaks at the start of each sentence to increase the time where needed.

    Login or Signup to reply.
  2. This is a SSML Text for the Azure api that works perfectly.
    The break tag must be used inside a voice tag.
    The time is defined in Milliseconds 1000 = 1 seconds

    <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="string" xmlns:mstts="https://www.w3.org/2001/mstts">
      <voice  name="en-GB-LibbyNeural">
    I have to think about it. Wait two seconds 
    <break  time="2000"/>
    Yes I agree  
    </voice>
    </speak>
    
    Login or Signup to reply.
Please signup or login to give your own answer.
Back To Top
Search