Whooshkaa's Text to Speech solution allows users to create audio from written text. Using synthesized speech, Text to Speech can create Media Library files or episodes on a regular schedule or one-off basis.
Note: Whooshkaa Text to Speech is available on all commercial Whooshkaa plans.
Text to Speech Documents and Jobs
Audio in Whooshkaa Text to Speech is created from Documents. Documents define the various parameters such as the content, schedule or the voice profile that is used for the speech.
When it's time to perform the synthesis, each Document will spawn a Job. Jobs are listed on the Text to Speech section in the Jobs tab and provide information about the current synthesis tasks.
Creating a Text to Speech Document
To create an audio from text, navigate to Text to Speech and click on the "Create a Document" button. Complete the following fields.
This field specifies the name to use for the document, in order to find it later in the Jobs or Documents.
Now or Scheduled
Choose between generating a one-off job now, or as a scheduled task. Scheduled tasks can be on a repeated basis, to be repeated daily, weekly fortnightly/bi-weekly (every 2 weeks) or monthly. The start time will be used as the basis for the first and subsequent jobs.
The resulting audio may either become a Media File or an Episode. The Title and Description entered will be used as the title and description if the resulting media. If you are choosing an Episode, you also need to choose the Show that the episode will go into.
If you are choosing the Media File option, you may also select the Media Files Library Folder to place into.
Note: The title and description also support Data Field templating (see below).
Whooshkaa Text to Speech supports one or more data feeds (JSON format) which can be used to place dynamic content into the output. At generating time, the data feed will be fetched and the data in the feed may be used.
The data feed has the following properties.
- Name: Used to refer to the feed in the listing
- Template Name: A name to refer to the feed in the content (must be unique)
- Data URL: URL of the data to use (must be publicly accessible)
- Format: Currently only JSON is supported
To reference data in the content, use the following syntax. For example, where a Template Name is "myData", the following can obtain the "quoteOfTheDay" key:
The text to be synthesized should go into the Content field. You may also use the templates from Data Feeds as per above.
Voice Character Settings
The content will be synthesized using the voice character. The characteristics such as the voice, volume, rate and pitch may be set.
To edit the Voice Character Settings, click on the title or the down arrow.
To preview the content in the configured character's voice, select some text and click on Preview.