Technology Innovations

Posts Tagged ‘Gesture Technology


Bangalore-based Dhananjai Bajpai, 24, is using the gesture-recognition technology to develop Kommunic8, a wireless wearable device that converts hand motions into speech. The device holds promise of improving communication for over 19 lakh speech-disabled people in India alone. A pilot project is currently being done with EnAble India, Bangalore, to test the functioning and accuracy of the device in the hands of actual users.

Dhananjai, who belongs to Kanpur city of Uttar Pradesh (UP), India, has completed his Bachelor’s in Electronics and Communication Technology degree from Shri Ramswaroop Memorial College of Engineering and Management, Lucknow, UP. He works at KFX Circuits and Systems in Bangalore and is also associated with Excubator, a startup incubator and corporate venturing advisory organization, where he works on Kommunic8 in his spare time. His aim is to use gesture-recognition technology for social benefit.

According to the 2011 Census of India, of the 268+ lakh total disabled population about 19+ lakh people suffer from speech disability. This population struggles not only in communicating with their surrounding environment, but also faces low job prospects that lead to another fight for quality sustenance. Kommunic8 aims to enable this populace to “talk” with anyone without any hesitation. 

Challenges

One of the challenges in bridging the divide between the speech-impaired and the common folks is that the sign language is difficult for a common man to understand and is restricted to the speech and hearing disabled community. Also, there is no standard international sign language that is followed consistently across the globe – each region and culture has its local sign language. These reasons prevent the differently-abled to communicate with others and live their social and professional life normally.

Kommunic8 equips the speech-impaired with a lightweight, wireless wearable ring-shaped device that can convert their sign language gestures into reasonable sentences in real-time and provide output in the form of an audible speech as well as a readable text on the K8 smartphone app.

With 97 percent accuracy and self-learning capability, the current prototype of Kommunic8 can be customized and programmed for any local language.

How it works

Kommunic8 is still in the development phase. Dhananjai began working on the technology in 2013, as a final year project, by creating a wearable glove which could detect the degree of bending of fingers and show respective alphabets on a mobile phone screen as per the American Sign Language. That was just the beginning which got its fair share of media attention. However, the wearable glove had its shortcomings in terms of size, speed, cost, and usability.

 

Initial prototype of Kommunic8

Initial prototype of Kommunic8

The current prototype of Kommunic8 uses a small circuitry packed neatly inside a ring like structure. The circuit uses gesture algorithm and a motion sensor that recognizes the sign and orientation of the user’s hand on which the device is worn. When the user moves or bends his/her hand to make a gesture, the sensor collects information and the software processes it to convert data into a sentence. The sentence is then spoken by a mechanized voice that is made audible through an inbuilt speaker. The same output can be presented in the form of a text on the K8 smartphone app.

The device operates on inbuilt battery that lasts for 10 hours – Dhananjai is working on increasing the battery power to last up to 24 hours. The device can be charged by any micro-USB charger.

The initial device will come with 50 actions predefined for ready use. The software, however, uses machine learning and will keep updating the database of gestures and sentences as the user starts using Kommunic8 regularly. Speaking of the storage capacity, Dhananjai says, “For now, Kommunic8 will come with a memory of 2 GB which can store up to 3,000 actions. This is sufficient, as on average a user may use maximum 100-300 actions in general. However, there is a provision for users to update the dictionary by connecting the device to a computer and make changes through the K8 desktop app.”

The device is supported by the K8 app available for Android and Windows phones that can be used to display the text, configure the device, recreate database and produce the speech output for interactions.  Here is a demo video.

 

 

Next steps

There is still a lot to accomplish before a market-ready version of Kommunic8 is complete. 

Dhananjai has filed a provisional patent for the technology innovation.

Now he is primarily focused on drastically improving the aesthetics of the device and using a more human voice output instead of the mechanized one. He is also working on including a small screen in the device with four push-buttons that can be used to reconfigure, expand and delete the database on-the-go, thereby removing the need for a secondary device for any kind of updates or assistance. 

Meanwhile, Dhananjai is reaching out to non-government organizations that might be willing to support pilot projects and provide sponsorship for further improvement of the device.

EnAble India is using Kommunic8 to:

  • Help teachers learn and improve their sign language and make classroom learning more attractive
  • Empower EnAble associated speech-impaired employees to use Kommunic8 for their daily communication at work.

“Results from the pilot will help me improve the device for the users. I am hopeful that Kommunic8 will allow them to get front-end jobs,” says Dhananjai.

Dhananjai can be contacted at dhananjaisrmgpc@gmail.com and +91-8765379454.

 

Advertisements

Chennai-based Mad Street Den startup has built a cloud-based platform that uses artificial intelligence to enable any smartphone with a camera to identify faces, detect facial expressions and emotions, and react to facial and head gestures. The expression acts as a trigger for a certain action, for example, frowning at the phone when an unwanted call comes in could make it shut down forthwith, or lifting an eyebrow could send the caller a message asking ‘What now?’

This image-recognizing platform, called MAD Stack, can be used by app developers and companies to create a futuristic mobile user experience. Mad Street Den founders Ashwini Asokan and Anand Chandrasekaran  explain that the idea is to make machines more useful by making them a bit human: fun, intelligent, and relevant.

The process of recognizing a human expression or responding to a gesture is simple for a human brain, but quite complex for a smartphone camera to do digitally. It is artificial intelligence that enables a camera to do this. What’s more, the app keeps getting smarter with use, through machine learning algorithms. Read more


Researchers at Microsoft and the University of Washington have found a way to harness electromagnetic radiation from the environment for a computer interface that turns any wall in a building into a touch-sensitive surface. Read more…


Enter your email address to subscribe to this blog's new posts by Email.

Join 76 other followers

Visitors Log

  • 10,446 visits till date

Monthly Archives

Advertisements
%d bloggers like this: