Enhancing AAC with AI: Sam Sennott Discusses Predictive Text, Voice Banking, and Cutting-Edge Technologies

The image displays the main page of an Augmentative and Alternative Communication (AAC) device layout for English speakers. It's a grid of colorful squares, each with a word or phrase and an accompanying pictogram to aid in communication. The squares include various categories like

In a detailed discussion on the transformative potential of Artificial Intelligence (AI) in augmentative and alternative communication (AAC), Sam Sennott, an assistant professor of special education at Portland State University, sheds light on how AI can significantly enhance the communication abilities of individuals with speech difficulties. Sennott, who has dedicated much of his career to AAC research, points out that the integration of AI technologies like predictive text into AAC devices is not a novel concept but has been evolving since the 1970s. These advancements have progressively improved the communication experience for users by leveraging AI innovations such as word prediction and dynamic word prediction models, which adapt to contextual factors like the time of day, location, and the user’s conversational topics.

Sennott emphasizes the substantial benefits of predictive text in AAC, particularly for individuals with slow typing speeds. By suggesting words and phrases based on the context of the conversation, predictive text not only speeds up communication but also introduces a dynamic element to it. This feature can significantly reduce the cognitive and physical strain on users, making communication less fatiguing and more engaging. For instance, it enables more dynamic storytelling in real-time, allowing users to express themselves more fully and naturally. However, Sennott also acknowledges the challenges associated with predictive text, such as the potential loss of personal voice and the cognitive load of evaluating predictive options, which can disrupt the development of motor memory and hinder the fluidity of communication.

The conversation also delves into the concept of voice banking, an AI-driven approach that allows individuals to record samples of their voice to create a synthetic voice that can be used in AAC devices. This technology is particularly valuable for individuals with degenerative conditions, enabling them to preserve and use their unique voice even as their condition progresses. Sennott highlights the emotional and psychological importance of voice banking, as it maintains the user’s identity and personal connection with their family and friends through the familiar sound of their voice.

Moreover, Sennott discusses the cutting-edge technologies in AAC, such as eye-tracking systems and noninvasive brain-computer interfaces, which have made communication more accessible and intuitive for users. These technologies, supported by advancements in AI, allow users to control devices and communicate through eye movements or brainwaves, bringing a sense of magic to the interaction. Despite these innovations, Sennott points out the disparity in access to these technologies, emphasizing the need for them to be more widely available and affordable, especially for individuals and communities with limited resources.

In conclusion, Sennott’s insights underscore the significant impact of AI on AAC, from enhancing the speed and dynamism of communication with predictive text to preserving personal identity through voice banking. The advancements in eye-tracking and brain-computer interfaces represent the frontier of AAC technology, offering new ways for individuals with disabilities to interact with the world. However, the challenge remains to ensure equitable access to these technologies, so that all individuals who could benefit from AAC have the opportunity to do so, reflecting Sennott’s broader vision for a more inclusive and supportive society.

Listen to the full interview below and hit the source link for an edited transcript.

Source: Marketplace

ChatGPT, a potential tool for increased accessibility, was used as a research and writing aid for this blog post. Do you think this is an appropriate use of chatGPT? Why or why not? Let me know!

Leave a comment

Your email address will not be published.


This site uses Akismet to reduce spam. Learn how your comment data is processed.