Google Aims to Help People with Impaired Speech to Live Independently

AI help improve communication options

The project is working towards improving the AI-based algorithms that are used for voice recognition and transcription programs so they can better understand words spoken by people with speech difficulties. So much of modern life relies on communication in one form or another.

People suffering from neurologic conditions such as stroke, ALS, multiple sclerosis, traumatic brain injuries and Parkinson's may have trouble communicating verbally. High-quality speech to text services might be one way to help improve their daily quality of life.

Speech recognition should help everybody

Project Euphonia is recording the speech of many people with limited speech abilities; initially, these are common phrases. These sounds samples are then converted into a spectrogram, or a visual representation of the sound. The system is then trained to recognize the less common form of speech of these highly used phrases.

Google wants to expand the algorithms' ability to recognize the speech patterns of people who speak other languages and may have a strong accent as well as people with speech patterns that are considered hard to understand due to cognitive impairment. ALS is a degenerative disease that affects more than 30,000 people in the United States.

Google Home may help people live alone longer

As the disease progresses most of muscle functions' loss makes it difficult to talk, swallow and breathe. People with ALS may find themselves becoming more isolated due to their inability to be verbally understood. The Google project may be one way ALS sufferer can continue to communicate.

The project can also help people live more independently with help from devices such as Google Home. Another goal of the project is to train algorithms to recognize sounds and gestures that represent commonly spoken demands that could be used in combination with Google Home.

Send your sample

This next step will be particularly useful for people with severely limited or no speech ability. Google is reaching out to people with limited or slurred speech to send voice samples to help them increase the sample size they have.

Larger sample size will help better train the algorithms. If you have slurred or hard to understand speech, then you can apply to send a sample by using this form.