Welcome to text2speachy! This is a text to speech device made to assist those who are visually impaired and struggling to read. The text2speachy device includes a speaker, a camera and many more parts working together to translate the text you provide into speech. It is designed to be easy to set up and use with parental guidance. This device is coded to take a picture of the text, process and convert it into speech at a rate that is understandable for young children.
Problem Statement
We focused on a topic for blind & low vision children, specifically those with inadequate access to technology and consistent power because we discovered that many visually impaired students struggle to comprehend braille, and some cannot read at all. After conducting some research, we determined that low vision students with access to audio devices have a higher graduation percentage compared to those who did not have access to those devices. Our objective is to market this device to younger children with limited access to electricity while inhabiting in middle to low economic environments.
Ideation Process
When planning our product, the first thing that we wanted to ensure was that this would be easily marketed and interesting to our audience, which is young visually impaired children. We considered things such as textile elements and a small friendly robot that would go on the students desk. This was later scrapped, but one idea that remained constant was that this should be simple, safe, and practical for young children. This ideal remained true after our conversations with some experts in the field. We received advice that pieces such as large overhead cameras and adjustable elements would be unrealistic for young children, so these were quickly edited and that can be seen in our final text2speachy design.
A similar ideal we had was to make this easy to use. Many devices similar to ours that are currently on the market are handheld or overpriced, we avoided anything handheld to make the use of our device simpler for our market, young visually impaired kids may find trouble locating the words on the page to scan so we landed on a stationary, non-adjustable device that scans the book when held in place.
Design Alternatives
When we were brainstorming ideas, a main point of difficulty came when deciding how the device would hold the kid’s book in place. Eventually we learned that having handles to hold the book in place wasn’t necessary and would simply get in the way.
1. No handles at all
- This was our fifth and final design and it was edited due to it being too difficult for the child to set up. In an interview with Dr. Sean Tikkun, an Assistant Professor in Curriculum and Instruction from North Carolina Central University
2. Sliders located on the sides of the base
- This was our fourth design and it is similar to our second design, but instead of the sliders located on the top, we moved them to the sides of the base to leave more space for the book. This allows the child to fix the placements of the handles easily, as well as easily flip the pages.
3. Adjustable hooks that fall into various slots
- This was our third design and it included 2 hooks that would smoothly fall into various slots on the base of the device. The issue was that this was the least adjustable option and got in the way of the book.
4. Sliders located on top of the base
- Our second design was customizable and versatile. We stuck with this design for the majority of the planning process until realizing that the sliding mechanism would easily get in the way of the child’s book.
5. A peg-board like base with adjustable handles
- This was our very first design and it worked very similarly to LEGO blocks, the design included two handles that would click into the board that had various points to receive the handles. Unfortunately, this was slightly flimsy and we became unsure of how the device would work.
Hardware: Chosen parts
After assessing all of our options for the camera, speaker, the text to speech module and our main board we decided to use a raspberry pi 4 type b, an Arducam MP5 ov5647 and an Adafruit mono enclosed speaker due to their efficient processing power and their compatibility with each other.
Design Criteria & Objectives
- Speaker that speaks 65 wpm
- Projects words clearly with adjustable volume
- Stationary with grips on the bottom so it doesn’t fall
- Mostly recyclable device
- Visually appealing
- Base target demographic ages 3-7.
- Electricity efficient
- Removable bottom to charge battery
Our objective is that we want to give kids who don’t have the resources to read braille a robot that can read books for them. A major reason we should be targeting this issue is because students with text to speech technology have higher graduation rates. Approximately 67% of students graduate from high school with this technology while 44% of students graduate high school without technology.
Design Evolution
Throughout the design process, our device was edited a lot seeing that our main goals were functional, easy, and electricity efficient.
- Prototype 1 is best be described as a robot that sits on the student’s desk. This was scrapped due to it being difficult for the camera to view the text.
- Prototype 2 looked similar to an overhead projector, this was removed after our conversation with Dr. Tikkun
- In our Final Prototype, we returned to a design similar to our first prototype but edited it so it would be simpler for the camera to scan the book provided.
Final Product
Dun Da Na Dunnnn… Here’s the final text2speachy prototype.
Result + Future Plans
In the future we hope to add several adjustments to our product in hopes to refine unsolved issues:
- A headphone jack to use in the classroom during tests or in other necessary scenarios
- The ability for the camera to register pictures and explain it to the kids
- A longer battery life or a energy source that won’t require electricity
- A more compact design fit for classrooms and small enough to fit in a backpack
- A human-like voice to improve the literacy rates
