The text-to-speech engine allows admins to create announcements from text quickly. Text-to-speech works by typing in text rather than uploading an audio file to the system. The system then generates an audio file that can be used in call flows.
While text-to-speech does not provide the quality of professionally recorded announcements, it does allow administrators to create time-sensitive audio, such as emergency announcements rapidly.
Use text-to-speech #
- Where appropriate tick the Text-To-Speech checkbox
- Expand the Voice dropdown and select either Female or Male
- MaryTTS ONLY: Under Pitch move the slide to define the degree of highness or lowness of the voice used for the announcement (mandatory)
- To raise the pitch (higher voice), move the slide rightwards (up the scale).
- To lower the pitch (lower voice), move the slide leftwards (down the scale).
- Under Tempo move the slide to define the speaking rate of the voice used for the announcement
- To make the voice faster, move the slide rightwards (up the scale).
- To make the voice slower, move the slide leftwards (down the scale).
- In the Text to convert to audio input field enter the text that you want converted to speech (up to 100 characters)
- Click the Play button to test how the audio sounds
- Click Add
- The audio will be created/uploaded
| Note: Text-to-speech may be powered by one of two Text-To-Speech (TTS) systems, depending on how Contact is configured at installation: either MaryTTS or Nuance. |
Why use text-to-speech? #
- Administrators can use the text-to-speech engine “offline” to quickly create announcements by typing in text rather than uploading audio files to the system. The system then generates audio files that can be used in call flows.
- Agents handling customer calls can use the text-to-speech engine during a call to read out booking confirmations, such as repeating back a customer number that has been entered through DTMF, or standard responses from a knowledgebase.