Link to Text Version Speech Research Lab
design bar
Info for Clinicians
Link to Speech Research Lab Main Page
Link to info for users of Augmentative and Alternative Communication Devices

Link to info for Clinicians

Link to info for Speech Researchers

Link to Model Talker Speech Synthesizer

Link to STAR : Speech Training, Assessment and Remediation

Link to Language and Speech Journal

Link to Speech Research Lab Newsletters

Contact Us!

 

 

 

 

 

Index:

General Information
About ModelTalker
ModelTalker Software
InvTool Software & Tutorial
BCCdb Software

Hardware Requirements
Inventory Design

General Information

The ModelTalker project is focused on the development of next generation text to speech (TTS) synthesis that is ideally suited for AAC device users. ModelTalker synthetic voices are ideal for AAC device users because they can be much more human sounding and more intelligible than older generation synthetic voices like DECTalk.

An important part of the overall ModelTalker project is the development of software that assists individuals in recording their own speech and then converts the recorded speech to a ModelTalker synthetic voice. This is a process often called "Voice Banking" and is of particular importance to individuals who are at risk to lose the ability to speak. Voices created with our voice banking software (called InvTool) can be highly intelligible and natural sounding and, crucially, capture recognizable voice characteristics of the person whose voice was recorded. Using InvTool an individual with ALS, for example, can record their speech, convert it to a synthetic voice for use in the ModelTalker TTS system, and then be able to use their personal voice in an AAC device when they are no longer able to speak intelligibly.

Development of software for the ModelTalker project has progressed to the point that we are now transferring the technology from our research laboratory to a small company that will turn it into a commercially available product. While this process is underway, we are continuing to allow people to download and use development versions of the software for free. As a clinician, if you have clients who might benefit from this software and are interested in assisting them in the process, we would be happy to hear from you and to answer any questions you might have regarding the project.

If you are discussing this software with clients who are considering using it, there are several factors you should keep in mind.

  • Above all, there are no guarantees! Users of this software must understand that they are not assured of being able to create a usable synthetic voice. We know that the software works reliably in our hands and it has worked well for others too (including ALS patients in their homes). However, some users were not able to create an intelligible synthetic voice or were not pleased with the voice they did create. If clients are uncomfortable with the possibility that they might spend 10 or 15 hours using InvTool and not like the result, this process might not be for them.
  • This software is still under development and as such, it is not a polished product with many support features. We do not recommend people use the software unless they are very comfortable with computers and computer software or have access to excellent local computer support.
  • While we value input from people who try our software, and always try to answer their questions, we do not have support staff available to assist you or your clients on a regular or extensive basis.