We were inspired by the fact that there are over 466 million people in the world with disabling hearing loss and globally the number of people of all ages visually impaired is estimated to be over 285 million. (WHO, 2018 & 2010, respectively) Yet, despite this pandemic creating numerous additional barriers for people with disabilities, there have not been many accessible, convenient, or affordable solutions.
Most people who are deaf and hearing-impaired depend on the ability to read lips to converse with others, and a facial covering that impedes communication can increase frustration and affect their mental health. When the volume of speech is reduced, the listener must concentrate harder to understand and follow the communication. Couple this reduction in volume with the inability to lip-read, and it can make it very frustrating for hard-of-hearing and deaf individuals, as well as the general population.
What it does
Our app Speech Assist helps people by transcribing speech into text on their screen in real-time. With masks being mandatory in many countries, its difficult for people with hearing impairments to interpret others. Our app provides real-time speech transcription of speech to text, text to speech, and video to sign language format. We also provide a resource and community help page.
The voice & speech recognition market will be worth $26.8 Billion by 2025 with speech recognition commanding the largest share due to increasing consumer demand, confidence, and usage. Over 466 million people have disabling hearing loss and over 285 million have a visual impairment. The speech to text API market size is 1.3 billion. This app meets consumers increasing demand due to most events moving online. Industries that will require it include healthcare, e-learning, and media & entertainment to optimize operations. Speech-to-text solutions offer faster transcription, low costs, and high accuracy. We could potentially sell access to the software like the API, and a Subscription for users to use the app, and ensure we can provide discounts for non-profits and support individuals if they cannot afford it, etc., This app has the potential to improve the lives of millions of people by Creating accessible platforms and safe spaces for people. We are leveraging growing technologies like AI, Blockchain, and Machine Learning to innovate our world. Voice and Speech recognition has numerous applications, including automotive, healthcare, education, hospitality, government, and public services, manufacturing/enterprise, retail, and e-commerce. Thus, the impact is infinite.
How we built it
Ethereum Web3.js, Truffle, Ganache, Solidity, HTML/CSS, Flutter, Dart, TensorflowLite, TeachableMachine
Challenges we ran into
Because our group consisted of members in the UK, India, Australia, and Canada, it was quite difficult to manage the time difference and continue the work through the midnight hours. Also, a lot of time was spent figuring out how to use the teachable machine and integrating the ML model in Flutter. However, we combatted this by staying in touch through the channels and keeping everyone in the know on what was happening and meeting virtually. In addition, Ideation was difficult as we had numerous ideas, and bringing together several ideas and possibilities down into a single app definitely took time. We went through various ideas and how we could execute them. In addition, finding the React component not working properly caused additional challenges to overcome and the integration was a bit messy.
Accomplishments that we're proud of
We are really proud to have accomplished the work that we had in the strict timeframe. Being able to learn from one another and communicate on a consistent basis was quite an accomplishment since we are all in different time zones.
What we learned
This app integrated principles including machine-learning, blockchain, and AI. Thus, it really was a learning curve and we learned new languages and frameworks. We learned a lot about the statistics related to people with disabilities as well as the progress and growth of the voice and speech recognition market. We also learned and developed time management and communication skills. It was a great experience to work with a group of inspiring women from across the world (for us, our group consisted of members in the UK, India, Australia, and Canada, and being able to manage the large time difference has proved that we are not only resilient but that barriers do not define us).
What's next for Speech Assist
The options and applicability of Speech Assist are endless, and we believe that this app will help numerous people globally. We created this app because we want to improve the lives of millions of people. We want to continue to create accessible platforms and safe spaces for people to support one another and improve communication across the community.
We would love to dive deeper and continue to leverage growing technologies like AI, Blockchain, and Machine Learning to innovate our society and advance our world. (Voice and Speech recognition has numerous applications, including automotive, healthcare, BFSI, education, hospitality, government, and public services, manufacturing/enterprise, retail and e-commerce, legal, consumer electronics, and military and defense), and technology type (artificial intelligence, and non-artificial intelligence) The options are endless for the different applications and subsets we can build with this app and structure.