What is ModelTalker?
The ModelTalker System is a revolutionary speech synthesis software package developed by the Nemours Speech Research Laboratory and designed to benefit people who are losing or who have already lost their ability to speak. It allows people who use a Speech Generating Device (SGD) to communicate with a unique personal synthetic voice that is representative of their own voice.
To learn more, see the FAQ section under Help, or select an option under our Voices menu to get started right away!
8/25/2016 – Email interruption
Some users are finding that email from modeltalker.org is being rejected as spam. This may impact your receipt of feedback on screening sentences or notice that a completed voice is ready for download. As a rule of thumb, you should have received a response to screening sets within three business days and if you have completed an inventory, you should receive a notice that it is ready to download within five business days (we typically do not respond on weekends). If you have not received a response within those time-frames, first check your spam/junk folder and if there is nothing there, please let us know so we can resend any communications.
6/21/2016 – New registration features
We have shifted to a new, more secure, password scheme. If you lose track of your password, it will no longer be possible for ModelTalker staff to send you the password. However, if you have forgotten your password, you can now reset it via a “Forgot password?” link on the login page. Along with this change, we have added an Account page to the Registration Menu that will allow you to change your password when you are logged in. Soon we will add some more features to the Account page to let you update email address and other information.
5/10/2016 – New Microphone Setting
We have introduced a new feature in the online voice recorder settings that shows which microphone our recorder is using. Always connect your microphone before starting Chrome, and always make sure that your USB microphone is the one showing in the settings dialog box. There is some additional help you can access from the Settings dialog if you have trouble.
1/21/2016 – Support ModelTalker
Our primary research efforts are directed to creating natural personalized voices for children who are dysarthric and unable to do standard voice banking. Your donation to ModelTalker through the Nemours Fund for Children’s Health supports that research, but it also helps us to continue to provide personalized voices to patients of all ages who have ALS/MND or another disease that can lead to loss of speech. If you or someone you know has benefited from our voice banking service, please show your support with a donation using the Contribute button on this page!
12/28/2015 – New Navigation and Registration options
The menus have changed on our website to improve navigation and make it easier for you to find what you need! We have also revamped our registration process. There are now three options for registering: Voice banking – the option for anyone who needs a personal synthetic voice for use with a Speech Generating Device or App; Voice donating – the option for people who want to donate a voice for others; Assisting – the option for clinicians and others who are assisting someone with voice banking. Unfortunately, we are no longer able to create free synthetic voices for any other reason.
12/21/2015 – New Features in the Web Recorder!
We have added a new feature to the web recorder that will make it much easier for users to review previous recordings and redo ones that had problems. In the Listen menu, choose recordings and you will see a list of all the recordings you have finished so far. Click on a sentence in the list and click the Listen button to review your recording of that sentence. Now you can also click the Record button to rerecord the sentence if you want to.
To further assist with navigating through the list of sentences as it grows long, the sentence number of each sentence is also displayed in the upper left corner of the center text display.
Important information on Logitech G430 microphones
We have recently encountered a number of users with Logitech G430 headsets who are experiencing problems with unacceptable microphone audio. This is a problem we have now also replicated in the laboratory on Mac computers, with the microphone input varying suddenly in amplitude. Our recommendation is presently to avoid purchase/use of the Logitech G430 for Mac users. We are not sure if this may also be a problem for Windows users, but it may be best to avoid that microphone headset entirely.
New Study on Speech Technology for Children with Cerebral Palsy
We are launching a new study as part of our Clinical Speech Technology Program! The goal of the study is to provide personalized synthetic voices to children with cerebral palsy and demonstrate the benefits to these children of having a voice of their own. For more information and details on participant recruitment, click here.
We are also actively seeking voice donors to build a synthetic voice with us that we can then personalize for a child with cerebral palsy to use in his/her SGD. If you live in the Delaware/Maryland/Pennsylvania area and have a child between the ages of 9-15 who would like to donate his/her voice, please email us for more information.
ModelTalker comes to the iPad
The UK-based company Therapy Box is working with us to bring this new technology to the forefront! Implementing ModelTalker voices into usable solutions is a brand new and unexplored field that has wider implications than someone just being able to use an app to communicate. Predictable 4 and ChatAble 2 will be the first iOS apps to utilize ModelTalker voices for AAC, greatly influencing the lives of those with communication difficulties. For more information about Therapy Box and these apps, click on the icons below.
Support for Mac OS X
You can now choose to download your ModelTalker voice for Apple computers running OS X 10.5 or newer. Your voice will be installed as a system voice that you can select from the System Preferences Dictation & Speech control panel.
Support for Android devices
ModelTalker voices are now compatible with the Android Speech TTS API. Your voice can be installed as the system default voice, and/or used by many Android Text to Speech applications.