Text To Speech Conversion Using Raspberry - PI
Text To Speech Conversion Using Raspberry - PI
Volume 4, Issue 2, February – 2019 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Text to Speech Conversion using Raspberry - PI
Vinaya Phutak, Richa Kamble, Sharmila Gore, Minal Alave, R.R.Kulkarni
Department of Electronics and Telecommunication Engineering
Abstract:- The TTS (Text To Speech) this technology is through vocal interface. Text extraction from colour image
basically used for conversion of text file into voice or in is very difficult task for computer the text to speech
audio form. This TTS technology proposed to help the conversion system is read the English alphabets and
blind peoples. According to world health organization numbers that are in the image using the OCR technique and
there are 91% peoples are blind. So there is necessary to convert it into the voice format. This paper presents the
help the blind peoples. For that purpose Text To Speech design implementation and experimental result of the
conversion using raspberry-pi proposed in this paper. It device. This device consist of two parts, image processing
includes camera to capture image as a input text, which module and voice processing module. The optical character
is then pass to TTS unit. TTS unit installed in raspberry- recognition (OCR) is the process that converts the scan or
pi and the output of TTS is amplified by using audio printed text images into the text format for the further
amplifier and then it given to the speaker. processing. This paper has presented the simple approach
for text extraction and its conversion into speech. The
I. INTRODUCTION testing of devices was done on raspberry pi module. Text to
speech (TTS) system produces the more natural voice that
This paper has represented the innovative idea as well can be closely matched with human voice. The example of
as low cost technique that is used to hear the contents of the the speech synthesis are the voice enabled e-mail and
text image without reading them. It combines the concept of messaging. The first step of speech synthesis is for the uses
optical character recognition (OCR) and Text to speech to speak a world in microphones and then that speech is
synthesizer (TTS) in raspberry pi. This system used to help converted into the digital format by using analog to digital
the blind people to interact with computer effectively converter and stored in memory.
II. BLOCK DIAGRAM
Fig 1:- Block Diagram
III. WORKING There are contain two concepts;
It can contain the following components: it is the OCR(Optical Character Recognition)
controlling peripherals. OCR is main element.It can convert the scanning
Power supply – It is an electrical device that can supplies image into editable text.OCR is implemented in this project
the electrical power.5v supply is used in that. to recognize characters which are then read out by the
Ethernet – It is used to access the e-dictionary website. systems through a speaker.
camera –This is used for capture the image
raspbeery pi –This project has been built around
Raspberry pi B+ processor board.
speaker –It is used to listen the text on the image.
IJISRT19FB152 www.ijisrt.com 291
Volume 4, Issue 2, February – 2019 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
TTS(Text to speech)
It is method that scans and reads english alphabet and
numbers. Webcam focused on text and it taking a
picture.there are some delay is required.after the delay,it
taking the picture and the processed by raspberry- pi
So that the steps for character recognition.
Webcam capture the image then image can read.
Pre-processing is done in 2nd step.colour image is
converted into gray scale and gray scale is converted into
the binary image.
Character is extract and resize the image.
Load templates that can be matched.
Remove the background
Edge detection is done in last step of character
recognition,in that open the text file and write the file.so
that the output is stored in text format. Fig 3
Python software is installed in raspberry pi.this
software is used to convert the image file to text file ny
extract the text.
IV. LITERATURE REVIEW
[1] reading is important in today's life. Printed text is
everywhere in the form of receipts, bank documents,
reports, books. There are many systems done text to
speech conversion, but they can't handle product
labeling. But big limitation is use of this is difficult for
blind people. This paper proposed camera based text
reading to help blind peoples to read the product label.
Main prototype of this system are 1) Novel based
algorithm to solve aiming problem. 2) Novel algorithm
of automatic text localization to remove the background
of the text image. 3) Camera based framework used for
bined people to read text.
[2] Chaw su thu et.al proposed text to speech conversion.
This paper considers computer based system that can
read any text,whether it was introduced in the computer
by scan and submitted to an OCR system .There are
many sysytems which convert text into speech.The OCR
Fig 2 system is used for the character recognition.The
recognize character save in notepad file as text.Then this
Above process is done in the image processing text file directly give to the computer as a input.Then
remaining process are done in voice processing.the image is speech through that using MATLAB.But this system can
convert into the text and text is converted into the speech. not detect small letters.
Following steps are required [3] Proposed that this project introduced smart device
In that espeak library is used. that useful for visually impaired peoples.This project
So that first we check the espeak library is available in uses the technology of camera based device that can be
computer or not.if the espeak library is not available then useful for blind people to read documents.The work is
we can get the error. implementing capturing image technique in an
Comparing the input string with espeak string embedded system.In this system has a camera as an input
Extract the voice device to detect the printed text for recognization and
that scanned text is process by a software OCR
Initialize the wave player for convert the text into
speech.
Finally we can get the output.
IJISRT19FB152 www.ijisrt.com 292
Volume 4, Issue 2, February – 2019 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
V. RESULT
Detect text on the image and convert it into audio file.
It can convert both capital as well as small letters.
VI. CONCLUSION
Text to speech can convert the text on image into sound. For
this conversion does not require internet connection. It is
very easy to use, so the blind person can independently use
this device.
REFERENCES
[1]. T.Rubesh Kumar, C.Purnima “Assistive System for
Product Label Detection with Voice Output For Blind
Users” International Journal of Research in Engineering
& Advanced Technology 2014.
[2]. Chaw Su Thu and Theingi Zin Implementation of
"TEXT TO SPEECH CONVERSION" international
journal of engineering research and technology volume
3, Issue 3, March-2014
[3]. Anusha Bhargava, Karthik V. Nath, PritishSachdeva
and MonilSamel “Reading Assistant for the Visually
Impaired” International Journal of Current Engineering
and Technology, 2015.
