Difference between revisions of "Text-to-Speech"

From Engineered Arts Wiki
Jump to navigation Jump to search
(Introduction)
(Introduction)
Line 8: Line 8:
 
= Introduction =
 
= Introduction =
  
Text To Speech,(TTS), converts text into sound.<Br>
+
Text To Speech,(TTS), converts typed text into sound.<Br>
[https://www.engineeredarts.co.uk EA] integrate third party TTS software to provide this functionality, most recent robots, 2014 onwards use [http://www.acapela-group.com/ Acapela].
+
[https://www.engineeredarts.co.uk EA] integrate third party TTS software to provide this functionality. From 2014 onwards EA robots use [http://www.acapela-group.com/ Acapela].
  
 
EA have also written software interfaces for:
 
EA have also written software interfaces for:
Line 16: Line 16:
 
*[http://neospeech.com/ NeoSpeech] - which has good support for Korean and Japanese
 
*[http://neospeech.com/ NeoSpeech] - which has good support for Korean and Japanese
  
All of the above TTS solutions are licenced, paid for software supplied by third party vendors, EA robots like [[RoboThespian]] and [[SociBot]] are normally shipped with at least one voice installed.<br>
+
All of the above TTS solutions are licenced, paid for software supplied by third party vendors.<br>
The default voice is the Acapela English male character Graham.<br>
 
  
Please specify any other voices you would like installed when ordering your robot, additional voices can also be installed remotely at a later date. There is a licence fee payable for each voice installed.
 
  
 
There are some free to use TTS solutions available like [http://festvox.org/festival/ Festival] these can be integrated by the customer if required but are not supported by EA.
 
There are some free to use TTS solutions available like [http://festvox.org/festival/ Festival] these can be integrated by the customer if required but are not supported by EA.
Line 77: Line 75:
 
</div>
 
</div>
  
== Installed Versus Virtual Robot TTS Voices ==
+
== Public User Interface Versus Virtual Robot TTS Voices ==
 +
 
 +
EA robots like [[RoboThespian]] and [[SociBot]] are shipped with a default Acapela English male character Graham voice installed on the public user interface [[http://wiki.engineeredarts.co.uk/Touchscreen#Compose Compose Screen]].  Additional TTS voices in multiple language and gender options can be ordered by customer at any time during or after purchase<br>
 +
 
 +
Customers using [[Virtual Robot]] automatically receive free access to 40+ Acapela TTS voices in a range of languages and genders to use in creating robot content.
 +
 
 +
 
  
 
== TTS Voices versus pre recorded speech ==
 
== TTS Voices versus pre recorded speech ==

Revision as of 13:56, 15 August 2017

Main text-to-speech.jpg

Introduction

Text To Speech,(TTS), converts typed text into sound.
EA integrate third party TTS software to provide this functionality. From 2014 onwards EA robots use Acapela.

EA have also written software interfaces for:

All of the above TTS solutions are licenced, paid for software supplied by third party vendors.


There are some free to use TTS solutions available like Festival these can be integrated by the customer if required but are not supported by EA.

Automatically generated mouth shapes and movements from TTS

Please note that Acapela Text-to-Speech (TTS) not only generates audio files from text, it also generates mouth shapes used by InYaFace on robots with a Projected Face.

Your projected face robot will lipsync with realistic mouth shapes and movements to any Text-to-Speech created with Acapela in Virtual Robot so you don't have to create the mouth movements yourself. You can modify these mouth shapes, or replace with your own in Virtual Robot if desired.

Technical details here - Text-to-speech Auto Generated Mouth Positions

This functionality is currently being developed to work with any Acapela voices installed locally on your robot as well.

Languages

TTS supports many different languages, and speakers, male, female, children, and characters.
There are also controls to vary the speed and tone of voices.

You test many of the Acapela voices on their website


As of Spring 2016 there are 34 Acapela Languages available:

  • Arabic
  • Catalan
  • Chinese Mandarin
  • Czech
  • Danish
  • Dutch (Belgium)
  • Dutch (Netherlands)
  • English (Australia)
  • English (India)
  • English (UK)
  • English (US)
  • Finnish
  • French
  • French (Belgium)
  • French (Canada)
  • German
  • Greek
  • Italian
  • Japanese
  • Korean
  • Norwegian
  • Polish
  • Portuguese
  • Portuguese (Brazil)
  • Russian
  • Spanish
  • Spanish (North America)
  • Swedish
  • Swedish (Finland)
  • Turkish

Public User Interface Versus Virtual Robot TTS Voices

EA robots like RoboThespian and SociBot are shipped with a default Acapela English male character Graham voice installed on the public user interface [Compose Screen]. Additional TTS voices in multiple language and gender options can be ordered by customer at any time during or after purchase

Customers using Virtual Robot automatically receive free access to 40+ Acapela TTS voices in a range of languages and genders to use in creating robot content.


TTS Voices versus pre recorded speech