Teaching Google to understand people with Down syndrome, one voice at a time.
The future is voice-first, but not for everyone. Because of their unique speech patterns, voice technology doesn’t always understand people with Down syndrome. Project Understood is ensuring the future of voice technology includes people with Down syndrome. The Canadian Down Syndrome Society is working with Google to collect voice samples from the adult Down syndrome community to create a database that can help train Google's technology to better understand people with Down syndrome. The more voice samples we have, the more likely Google will be able to eventually improve speech recognition for everyone.
“My friends can understand me.
Why can’t voice technology?”
-Joshua, Participant
Add your voice to Project Understood.
Machines learn through data. The more data they get, the more accurate they are. To teach voice technology, the data we need are voice recordings. The Canadian Down Syndrome Society is partnering with Google to give them the voices they need to better understand people with Down syndrome. The more voice samples shared by the Down syndrome community, the closer we get to a world where every person is understood.
step 1
Click ‘Donate Your
Voice’ to complete
Interest Form
Voice’ to complete
Interest Form
step 2
Receive Email Login
from Google
from Google
step 3
Start Recording Phrases. (See FAQs for more info)
step 4
Share on Facebook &
Recruit Others with
Down syndrome
Recruit Others with
Down syndrome
step 5
Each Voice Added
Will Help Google
Will Help Google
“There will be 8 billion digital
voice assistants in use by 2023.”
-TechCrunch1
The Technology
Voice interfaces have now been sold in millions of products ranging from smartphones, to vehicles, to home devices. These systems offer endless possibilities for enhanced living. But as it currently stands, the technology is not optimized for use by people who would benefit from it the most: people with disabilities. Automatic Speech Recognition (ASR) can greatly improve the ability of those with speech impairments to interact with everyday smart devices and facilitate more independent living. However, these systems have predominantly been trained on ‘typical speech’. But not all human speech is the same.
The unique speech patterns of people with Down syndrome make it difficult for voice technologies to understand them. This is due to a large lack of training data. The Canadian Down Syndrome Society along with Google’s Project Euphonia are setting out to make speech technology more accessible to those with disabilities by recording the voices of thousands of participants with Down syndrome to help train and improve its technology.
By reading and recording simple phrases, we can help Google recognize your unique speech patterns to improve Google’s system. Your voice recordings will be used for the purpose of research, and with the goal to ultimately improve the accuracy of speech recognition for people with Down syndrome. With your help, we can create a world where people with Down syndrome are better understood.
DONATE YOUR VOICE
The unique speech patterns of people with Down syndrome make it difficult for voice technologies to understand them. This is due to a large lack of training data. The Canadian Down Syndrome Society along with Google’s Project Euphonia are setting out to make speech technology more accessible to those with disabilities by recording the voices of thousands of participants with Down syndrome to help train and improve its technology.
By reading and recording simple phrases, we can help Google recognize your unique speech patterns to improve Google’s system. Your voice recordings will be used for the purpose of research, and with the goal to ultimately improve the accuracy of speech recognition for people with Down syndrome. With your help, we can create a world where people with Down syndrome are better understood.
Examples of phrases
you’ll be asked to record.
“We’re committed to bridging
that gap and making it
work better
for people with
Down syndrome.”
– Julie, Product Manager at Google
“For the rest of us it's
kind of fun,
but for him,
it's necessary.”
– Joshua’s Mom, on voice technology
VOICES DONATED
Each voice helps bring us closer to our goal of making sure more people with Down syndrome are understood.
0
500
1000
FAQ
Project Understood is a partnership between CDSS and Google. The goal of Project Understood is to collect voice data from adults with Down syndrome to help improve Google’s speech recognition models, specifically for that group.
The Canadian Down Syndrome Society is partnering with Google to make voice technology more accessible for individuals with non-standard speech. One of the challenges for Google has been recruiting enough people to participate in the data collection process. CDSS saw this as a great opportunity to collaborate and offer our help in reaching individuals within the Down syndrome community. Between Google’s technological expertise and CDSS’ connections, this partnership works to further Google’s research. Additionally, CDSS is working to ensure that individuals with Down syndrome are being well represented in the future of voice technology.
In order to apply to participate in this research initiative, you must meet the following criteria:
- Be at least 18 years old.
- Have access to a computer with a microphone or a mobile device (Android or iOS).
- Speak English fluently.
- Have difficulties being understood by others (and not just because of an accent).
After you submit an interest form, you’ll receive an email from Google’s team within 3-5 days. This email will provide you with login details for ChitChat. ChitChat is the program you’ll use to record voice samples. In ChitChat, you’ll see a list of test phrases. You can hit the record button under each sentence to record yourself reading the phrase, and then save it. This first set of phrases will help us determine if you’re eligible to continue. This screening process helps us to ensure that there are no technical issues and that the program is ready to hear your voice.
If eligible, you’ll receive an email with a link to the full sample set of 1700 phrases to record. You don’t have to record all of them at once and can take as much time as you need to get through the full list of phrases. However, we suggest trying to finish them over the course of 1-2 weeks. Once complete, Google’s engineers will process the data and use it to help voice technology better understand people with Down syndrome.
Voice technology is what lets you speak with digital assistants on your phone and in your home. It uses voice recognition to turn spoken language into text so that it can be understood by a computer. Google has been at the forefront of this field in recent years, and with initiatives like Project Understood, they're working to make it accessible to more people.
We understand that recording these samples will take time and energy. As a token of appreciation, Google will provide thank you gift cards to participants who share their speech samples with us.
Your voice data will be used to improve Google’s speech recognition models. Only Google employees and affiliates will have access to your recordings. Your recordings will not be available to anyone else unless you give Google explicit permission to do so.
We know that current speech recognition technology can be frustrating for individuals who have difficulty being understood. However, Google is still in the early stages of a potentially long research arc. We hope our research leads to product improvements as soon as possible, but realistically this may still be a few years out. More voice samples will accelerate the research timeline.
To contact the Canadian Down Syndrome Society, please send email inquiries to info@cdss.ca.