AIHEARU

A Case for the Cloud

Arun Munje, CEO

What was the business problem?

Speech-to-text technology has really advanced over the last 20 years, and for many people it’s now the go-to method of interacting with their devices. But even as the technology’s improved, it’s really only effective for people whose voice and speech patterns are similar to those of the general population. Currently, speech-to-text uses crowdsourced data to train its artificial intelligence (AI) algorithms to recognize what is being said. People with cerebral palsy, stroke victims, and others with speech problems can’t use speech-to-text since the crowdsourced speech doesn’t sound like their speech. The accuracy of speech-to-text for those with speaking problems can be less than 10%. If speech-to-text is to become truly universal, it has to be able to support all speakers, including those who have difficulty speaking.

How does AIHEARU solve this problem?

AIHEARU is a speech-to-text system that learns a user’s unique speech patterns over time. The AI in AIHEARU uses a foundational data set for a user’s initial interaction; however, the more a user interacts with AIHEARU, the more it becomes tailored to their unique way of speaking. Given time to learn, AIHEARU boasts an accuracy rating of over 90% for individuals who have difficulty speaking clearly. Along with the AI model, AIHEARU offers an open API that allows integration with popular communication platforms such as Zoom, or voice assisted services like Siri, Alexa, and Okay, Google. Integrating AIHEARU with existing voice-based communication systems could expand the market for many services currently inaccessible to those that have difficulty speaking.

What technical resources did you need?

We were searching for a based-in-Canada cloud provider that was able to offer us enough power to build our AI model, while also remaining affordable. The DAIR Cloud Program provided that and more. The access to a Graphics Processing Unit (GPU) was also a nice bonus.

Why did you choose the DAIR Cloud?

I had experience using DAIR when working for a previous venture that I co-founded, named Vayyoo, so when I was looking for a cloud provider based in Canada, I knew DAIR was there to help me build my business. I was pleasantly surprised to also be able to access a GPU for development, as this extra power was critical to developing and training our AI model quickly.

How did the DAIR Cloud help your business?

The DAIR Cloud provided the resources we needed to run the development environment of AIHEARU, while also running a production environment for our beta testers.

Beyond the cloud resources, we received great technical support from the DAIR team, the DAIR Slack Channel, DAIR’s regular webinars, and archive of materials from earlier presentations. All of it was very handy.

What was the result?

AIHEARU is now in beta testing with a small, dedicated group of users from many different parts of the world. We are in the process of developing our own voice assistant named “Buddy”, that will further enhance AIHEARU’s capabilities. We’re also continuing to develop our API to work with even more services and platforms.