Using Text to Speech

Course text and simulation step text can be converted into speech. Text to speech provides accessible outputs for visually impaired users. Authors can automatically convert text to audio that is available to the end user when playing the published output. In addition, text to speech allows audio to be translated and used in conjunction with XLIFF files.

Configuring Available Voices for Text To Speech

1. Ensure that text to speech is enabled for simulations and courses in the template.

   NOTE:

   For more information on enabling text to speech, refer to Default Publication Settings: Simulation: General Settings and Default Publication Settings: Course: General Settings.

2. Configure a voice for text to speech.

   NOTE:

   For more information, refer Specifying General Settings: Text To Speech.

   NOTE:

   The MS Speech Platform and any applicable voices must be installed on your local computer.

3. From the Editor, open the course or simulation for which you want to add text to speech.
4. Select File > Document Properties....
5. Click Text To Speech on the left pane.

1. Select one of the following options:

   If You Want To	Then
   Use the default template voice	Select Use default template voice.
   Select a different voice	Deselect Use default template voice. Select a voice from the Available Voices drop-down list. Enter text in the Preview Text field. Click Play to preview the selected voice. Adjust the Volume and Rate as needed. NOTE: Click Reset to return the volume and rate to the default settings.

2. Click OK.

   NOTE:

   Text to speech audio is published within the course or simulation.

Text-To-Speech Voices Provided by AWS Polly

Requirements:

• uPerform Authoring Client 5.43.111 or higher
• This feature is an add-on, enabled by product keys. Please contact Customer Success to obtain a new key.
• This feature requires an active connection to a uPerform Server
• This feature requires specific settings to be enabled in the uPerform Template
• "Store Generated Audio In Documents" must be enabled and the template saved after enabling the option
• Text-To-Speech generation will be performed during a SAVE operation on a document. 

Restrictions:

• Maximum of 3000 characters PER AUDIO FILE
  • This means that each step or stencil can support a max of 3k characters for that step. 
  • This does NOT Mean a max of 3000 for the entire document.
• No control over voice Speed or Volume

Rate Control

Rate control is available in the template and in individual documents and courses.

The minimum value for rates is 20 and the maximum is 200. If nothing is selected or an out of range value is entered, the rate will default back to 100. 

SSML Transcript Support

1. Go to audio tab (Standard options are still available).
2. Go to the transcript window and enter SSML. SSML is only available inside of the transcript and putting SSML anywhere else will not work.
3. After placing SSML in the transcript text area, you’ll see a red warning appear above the transcript suggesting that you test your SSML by pressing the play button to ensure it is valid. If invalid SSML is entered, the system will generate an error warning to notify you that your SSML is invalid.

There is no limit on the number of tags. The tags must use the valid SSML supported format and contain the value. The only supported SSML tags are listed on the AWS support site.

Long-form SSML (currently in Beta via AWS) is not supported at this time.