Hand written text to speech
The idea is that blind or visually impaired persons who can use a smartphone can hear the speech of text that they want to read. By using our application, the user needs to upload the image of the text, either by capturing it or importing it from the smartphone gallery. Then the application will per
2025-06-28 16:32:50 - Adil Khan
Hand written text to speech
Project Area of Specialization Artificial IntelligenceProject SummaryThe idea is that blind or visually impaired persons who can use a smartphone can hear the speech of text that they want to read. By using our application, the user needs to upload the image of the text, either by capturing it or importing it from the smartphone gallery. Then the application will perform the following steps:
- Capture image: Images will be captured from the environment by the smartphone camera. The image will be converted into a gray scale through the image binarization process
- Process data: Received data will be processed by using deep-learning techniquesSegments will be created from the grayscale image through an image segmentation process. The text will be recognized through the convolutional neuronal network
- Send a speech message to the user: Speech message will be transmitted to the user. The recognized text will be displayed on the display panel and simultaneously converted to voice using an Android function called speech synthesis. The voice will be heard by the user.
Therefore, the application is simple and easy to use for blind or visually impaired people, as well as for anyone who understands English.
Blind assistant is an android based application that recognizes the Hand-written text of the user who is visually impaired or blind. This application will give assistance to visually impaired or blind people to convert text into speech. Visually impaired people are not able to read correctly the simple warnings in walls or signals that surround us. Thus, the development of a mobile application that can perform the image to speech conversion has great potential and utility. We will accomplish the task with the help of machine learning algorithms applied to the challenging problem of the visually impaired person.
Project Implementation MethodWe will utilize the geometric features of a handwritten letter by using an artificial neural network, then implement by OCR (Optical Character Recognition). Our goal with a blind assistant is to make an application which can fit generally in similar kinds of problems. The Android studio is Mobile application software which is a modified form of Linux and other open source software. We will be working on those programming languages which are used in android studio for the purposes of making an application.
Benefits of the ProjectBlind Assistant is to be made for those people who are visually impaired or facing difficulties to read English, It allows people to get the information they want without a great deal in smartphones or tablet. This application is convey the hand-written text which written on the paper in the form of speech to assist the visually impaired person who can easily understand the content of an image. It aims to help the maximum number of learner comprehend and retain information by appealing to learning styles. It gives a chance to listen to the content instead of reading it allows people to do something else at the same time. This application serves as a means of communication between human and machine. This application will be useful for visually impaired people who want to read the handwritten text in English without human assistance. It will play an important role in the country's economy by helping the blind towards education.
Technical Details of Final DeliverableBlind Assistant is an android based application which will guide blind people outside well-known environments. Our system will be composed of two interconnected parts which will run concurrently. A handwritten text-to-speech system is composed of two parts i.e. a user interface (front-end) and a background functionality (back-end). The objective of this project is to identify handwritten text with the use of neural networks. The interface performs two major tasks. First, it converts raw text containing symbols like numbers and abbreviations into the equivalent of written-out words. We must construct suitable neural network and train it properly. The program should be able to extract the characters/words one by one and map the target output for training purpose. After automatic processing of the image, the training dataset must be used to train “classification engine” for recognition purpose. Second, it assigns phonetic transcriptions the speech synthesis module to each word, and divides and marks the text into prosodic units. In background, it converts the symbolic linguistic representation into sound.
Blind Assistant shall be installed through the .apk file. The .Apk file allows you to install an application wherever you want.
Final Deliverable of the Project HW/SW integrated systemType of Industry Education , IT Technologies Artificial Intelligence(AI)Sustainable Development Goals Quality EducationRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 31000 | |||
| Printing Material Cost | Miscellaneous | 1 | 4000 | 4000 |
| Graphics Material cost | Miscellaneous | 1 | 6000 | 6000 |
| Echo Smart speaker | Equipment | 1 | 11000 | 11000 |
| Deployment of application on google play store | Equipment | 1 | 10000 | 10000 |