Hey, I am working on a project which involves scanning a text and the exact text can be heard via the speaker. The only thing I only need to use is Arduino.
How can it be done, I mean are there any already existing projects that can do Object Character Recognition using Arduino ??
and some sort of optical sensor?
...and you also need the ability to generate audio, maybe an audio amplifier, and a speaker.
how to achieve it that's what I need plz help
What is your level of experience with Arduino, electronics and coding?
Why the constraint on only using an Arduino ? It seems to be an unlikely choice given the nature of the project
Sir, if you were the head of this project what would you have preferred instead of arduino
I have been working on Arduino-related projects for 1 year.
A PC would be preferable to an Arduino and you obviously have one
The thing is I need to use a development board for this project which is y I m after Arduino
I presume that you have done a Google search for "arduino ocr" and a search of this forum
Yeah, and no good answers.
Is the text in a known font? Preferably OCR-A?
Memory is the limit. A full page scan at 300DPI is gonna be huge (90k/sq in) Since you're reading aloud you can scan a line or two at a time instead of the full page and OCR as you go. You can easily detect line breaks IF the page isn't skewed going thru the line scanner.
Reading is a whole 'nother issue. Storing a library of individual words in .WAV or .MP3 files is going to take resources that Arduinos aren't really appropriate to use.
A better bet might be a phonem based approach. The old Votrax SC-01 chip knew them and you just printed to it in a corrupted ASCII-sorta code.
Problem is the to get natural sound you need a lot of tweaking and you still wind up sounding like Stephen Hawkins' VOX.
What format is the text in? Is it complex like a standard sheet of paper with 20+ rows of text and embedded graphics, or something much simpler like a continuous stream of characters (think old fashioned ticker tape).
"Exact" can frequently cost a lot of money and effort to achieve.
I would probably be looking at a Raspberry Pi solution, maybe with existing open source OCR software. If you're feeling much more adventurous, then you might be able to use an ESP32-CAM board and port an OCR library to it.
It's just normal and small words like
A for apple
B for boy ........
Use a Raspberry Pi with camera and audio amplifier.
This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.