Jaydeep Solanki and team have come up with a free app which enables the visually challenged to read text using a phone camera. HT's Rahat Bano engages them in an exclusive chat. Here are few excerpts:
Tell us something more about the Read For Blind project. How did the idea of an app for the visually challenged strike you?
Actually it wasn’t my idea. A team member (Abhinav Tripathi) was with his father at a cancer hospital, beside which was a blind school. He saw students struggling to study. He told us about this, and thus we decided to do something about it, meanwhile we heard that Microsoft Imagine Cup (IC) was coming, so thought it would be great to showcase our skills and innovation.
What were the most challenging parts of the project? What was the journey to the final outcome like?
The most challenging part was to implement image processing algorithms; we had to code it from scratch as there was no image processing library available for Windows Phone 8 (WP8) that would help us.
What kind of tweaking did you do to the app during and after the pilot?
Before IC it was all like code, code, code. The main aim was to improve the app and make it cope up with the Microsoft standards.
How has it been received by different people? What kind of a response did you receive from the participants/volunteers during the pilot? What about industry? You took part in the Microsoft Challenge, where the app won the third prize.
Other participants were amazed to see an app like this, maybe it was the first time they were seeing a phone taking decisions by itself. We did a test in a blind school, and it went really well. Yes, we won the 3rd prize.
What kind of queries and feedback did you receive there?
The feedback was marvellous. The judges were happy to see the youngest participants (us) develop wonderful projects. They gave our app some stuff to read, and it read all of it correctly. They were impressed!
You have made a free app. What prompted you to do that?
We think that a blind person doesn’t earn too much to take away. So making this a commercial app won’t serve its purpose. Moreover, we built it keeping the blind students as our target users and we personally think that students should be provided study aids for free.
At what stage is your project now? Is the app available on the app store already?
Currently our app runs on high-end devices only, we are trying to optimise it to make it work on entry level devices, too, because obviously a blind person won’t spend money on a high-end device. No, the app is not available on the marketplace, because we think that it still needs a lot of fixes.
What are your learnings from it? With what expectations did you start this project and how do you think it has turned out?
We learnt a lot of great image processing algorithms. We started the project with an aim to help blind students read independently. We can say we are almost on the finish line, just a few more steps and we are done. It turned out really well, more than my expectation.
I remember the days when we were developing an app for WP8 and none of us had a WP8 device. We tested it using the Emulator, and now I own Nokia Lumia 920 (thanks to IC).
Apart from possibly the Hindi speech feature, how is your app different from other such developments made elsewhere?
Our app lets a blind person click a photo on his own. It guides him as to whether the all text is within the frame or being clipped.
What are you doing now? (Which course and semester are you in? What do you plan to do after graduation?
I’m pursuing BTech in information and communication technology (ICT). This July I’ll start with my fifth semester at DA-IICT. After graduation I would like to take a higher degree in computer science.
Tell us something about yourself? Why did you get into engineering? Why DA-IICT - what’s it like being there?
When I was in 12th standard I knew nothing about engineering, but I knew one thing, that I love coding stuff, plus somewhere inside I wanted to work with electronic circuits, too. Seeing DA-IICT’s curriculum I knew this was the institute for me! That’s how I ended up at DA-IICT. DA-IICT is like a dream institute. It gives you everything you need. What I enjoy the most here is freedom and the learning atmosphere.
Team: 4 members - leader Jaydeep Solanki (sitting in the photo), Somsubhra Bairi (in a blue T-shirt), Abhinav Tripathi (brown pants), Anuj Kosambi (plus a classmate, Nainesh Iranaya, who helped the DA-Developers team with a Windows phone, etc)
Focus Area: People with disabilities/mobile technology
Outcome/impact: A free app
What next: Trying to make it work on entry-level devices, too
‘The students wanted to do something tangible for society’
Asim Banerjee, a professor at the DA-IICT, mentored the students in their quest for a solution to help visually challenged people read on their own “to some extent”.
The app, voice controlled so the user can navigate through it, converts images to text and then text to speech. The user needs to place his phone over the text, such as a book, label on a packaged product, or business card, to be read and the phone guides him/her as to whether the required text is within the frame. The phone captures a photo which is converted to text, using OCR (Optical Character Recognition) service of Microsoft Hawaii. The app reads out the text to the person. The team’s aim was not to “launch it as a product, but to give it as a free app”. While the app currently has only the English speech feature, the team is now working to make it read more languages, says Banerjee. “It’s almost ready for Hindi and Gujarati.”
Banerjee points out that the project is not over yet. They plan to continue the work to improve the app. “We intend to take this further. We would want to do a detailed user study, enhance the application by adding features (like multi-language support) that we think would further improve it — in terms of its usefulness to the target audience based on the feedback we would be receiving from the users.”
Talking about the team’s evolution, he Banerjee, “I would say that in the initial stages the team was a group of young minds bubbling with enthusiasm, with a strong desire to learn and do something tangible for society. They had a lot of ideas — but in my opinion — the differentiator of this group vis-à-vis others in the same age bracket, was a desire to implement their ideas. They were willing to harness whatever they knew into fructifying their idea...and also willing to learn new things along the way whenever and wherever required.”
Initially, he says, the students unquestioningly accepted suggestions as “gospels” but later into the project, they began to expressing their views politely.
“I think they have learnt a very important lesson of voicing their concerns and reservations without offending others,” says Banerjee.
Five facts at your fingertips
1. About DA-IICT Dhirubhai Ambani Institute of Information and Communication Technology (DA-IICT) was founded in Gandhinagar in 2001. It became a university through an act of the Gujarat legislature in 2003. A private non-affiliating university, it does not take any aid or other financial assistance from the Central or state governments
2. Flagship programmes DA-IICT offers a four-year undergraduate programme leading to a BTech (information and communication technology) degree. The programme is designed to prepare students for either a professional career immediately after graduation or to take up postgraduate studies in India or abroad, says the institute
3. Placements: The average pay packages of DA-IICT students were Rs. 4.46 lakh per year in 2012, Rs. 3.98 lakh per year in 2011 and Rs. 3.61 lakh per year in 2010 while the highest packages were Rs. 16 lakh a year, Rs. 8.5 lakh per year and Rs. 7.5 lakh per year, respecively. The companies which recruited DA-IICT students include Google, Microsoft, IBM, JP Morgan Chase, and others
4. Gandhinagar The capital of Gujarat, it is the twin city of Ahmedabad
It is home to reputable institutions - the National Institute of Design, National Institute of Fashion Technology, Institute for Plasma Research and others
You can find a lot of eating options
5. Admissions DA-IICT’s UG student intake for 2013 is 300. The tuition fee for 2012 was Rs. 40,000. Generally, the sale of application forms for BTech begins in mid-March and ends in late April. Last year, enrolees’ AIEEE ranks ranged from 1596 to 14,972 (i.e. lowest rank was 1596 and highest, 14,972) in the all-India general category. More at daiict.ac.in