Skip to content

A collection of pre-built speech synthesis settings used to convey emotion

License

Notifications You must be signed in to change notification settings

brycehowitson/SSML-prosody-library

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 

Repository files navigation

SSML-prosody-library

A collection of pre-built speech synthesis settings used to convey emotion.

Prosody "presets" (Volume, Pitch, Rate) matched to theatrical voice descriptors, including terms like these:

  • grumble
  • ambivalent
  • harsh
  • sarcastic
  • scornful
  • arrogant
  • impassioned
  • indignant
  • inquisitive
  • sincere
  • jovial
  • solemn
  • thoughtful
  • elated
  • threatening

Here's a little background on the project and rationale on Medium


Example of "Excited" Prosody

<prosody rate="fast" pitch="high" volume="loud">This voice is very excited!</prosody>

SSML can be tested using IBM's interactive demo


How to contribute

Any help would be much appreciated! I'm no expert in the human voice but this library may prove extremely useful to anyone developing a VUI tool.

Simply add an xml document to the repository containing the prosody tag and attributes to emulate a style of voice. Make sure to name the document with your term. For example the above lives in excited.xml.

About

A collection of pre-built speech synthesis settings used to convey emotion

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published