I love this wave of new technology that we are riding, especially when it brings us new innovations that really change lives. Two up and coming Augmentative and Alternative Communication (AAC) developments are great examples of this, and they both really have me excited about their potential!
ModelTalker (MT) “is an engine for creating voices from recordings of a patient’s natural voice.” It allows individuals who are losing their speech, due to disorders such as Amyotrophic Lateral Sclerosis (ALS), to record their voice while it is still understandable. These recordings are then transformed into a unique, synthetic voice to be used later in an augmentative communication app. MTs demo page allows you to sample this technology by typing in phrases and having them spoken aloud using synthetic voices developed from three different speakers.
In order to create a voice, you register with ModelTalker and record preliminary sentences with their web-based recording tool or with their downloadable recording software. Once that is approved, you will receive an entire inventory of phrases that will take 4-6 hours to record. When finished, ModelTalker will create your synthetic voice and send you a link to download the voice to your device.
Therapy Box is working with ModelTalker to incorporate this speech technology in their new updates of Predictable 4.0 and ChatAble 2.0, which will be coming soon. Although ModelTalker has been available in some form for many years, this will be the first time that it will be user-friendly, accessible and easily implemented within a communication app. This will enhance the practice of voice banking and allow individuals to say precious novel things to their loved ones using their “own” voice.
TalkItt is a communication app that translates unintelligible, but consistent speech into an understandable voice. It works similarly to Google Translator, where you speak a sentence into your phone or iPad in one language and out pops the same sentence in a different language. The difference being that TalkItt needs to learn each user’s speech patterns to create a personal dictionary. The app will then pull from this dictionary every time the user speaks into the microphone. For example, if the speaker says, “ilaayoa”, the app would translate with information from the user-created dictionary and speak in a synthetic voice, “I Love You.” This video is a good example of how it works. See all of their videos on VoiceItt’s video channel. Although the exact specs of this app have not been finalized, it is anticipated by the company to run on many mobile and wearable devices.
I am particularly excited about this technology, as are some of the people I work with. To my knowledge, it will be the first AAC application that will be activated by a user’s voice versus touching a screen, series of buttons, or a switch. Without having had the chance to try it, in my opinion, it has the potential to ultimately improve speed and access to voice-output communication for people with speech disorders.