• Using Python & PyQt machine learning and natural language processing, we developed
software that can translate videos from English to Hindi automatically.
• Using Google Translator and Python tools, increase translation accuracy by using the
splitting methodology.
1. B. P. Mandal College of Engineering, Madhepura
(Dept. Of Science & Technology, Govt. Of Bihar)
Major Project
Speech Dubbing Software
Presented By:
Pushkar Kumar
Kapil Kumar
Pranav Ravi
Sudama Manjhi
Project Guide:
Mr. Raj Kumar
Assistant Professor
Dept. of CSE
@copyright by Team Ciphers-BPMCE Madhepura
2. Contents
• About Project
• What is Dubbing?
• Block diagram of Project
• Project Dubbing Algorithm –I&II
• Divide / Conquer Logic of recognition
• Synchronization of Audio
• Project Flow Chart Diagram & Working Methodology-I&II
• Used Language , Tools & Environment Description for Project Implementation
• Project Dashboard Screen
• Future Scope
• Conclusion
3. About Project
It is a Machine Learning technology &Natural Language Processing
based project. In this project we developed a GUI based software. The
function of this software is to convert the .mp4 English to .mp4 Hindi
Video automatically.
Speech Dubbing Software
.mp4_English
.mp4_Hindi
4. What is Dubbing?
Dubbing is a post-production process where the original language of recording
is swapped with audio in a different language and is then mixed with the audio
of the media to make it sound as natural as possible.
Original language
Audio
Swapping
Desire language
Audio
Mixing
Desire language
with File(.mp4/.mkv)
6. Project Dubbing Algorithm -I
Step 1: Take input video file from
desktop to program for dubbing
Step 2: In this step we extract the audio from the input
Video file and stored in directory as generatedSpeech.wav
Step 3: Again we import the generated Speech.wav from directory
and divide the audio into multiple parts.
Step 4: In this step we recognise the text of each audio clip and stored in
directory as generatedSpeech.txt
7. Project Dubbing Algorithm -II
Step 5: import the generatedSpeech.txt from directory and
translate the file English to Hindi and Stored in
translatedHindi as a variable.
Step 6: Now we generate & synchronized the audio file from
translatedHindi and stored in directory as translatedHindiVoice.mp3
Step 7: To overlapping input the video file with translatedHindiVoice.mp3.
& Stored in Directory as convertedOutput.mp4
Step 8: Remove all unnecessary generated files like GeneratedSpeech.wav,
generatedTxtFile.txt, TranslatedHindiVoice.mp3
8. Divide /Conquer Logic of Recognition
Total duration :17 min
60 sec
Total No. of frame =17
Audio File .wav
Speech recognition
generatedSpeech.txt
12. Used Language , Tools & Environment Description for Project Implementation
Pycharm: PyCharm is an integrated development environment used in computer
programming, specifically for the Python language. It is developed by the Czech
company JetBrains.
Qt Designer It is the Qt tool for designing and building graphical user interfaces
(GUIs) with Qt Widgets.
Python: It is a powerful general-purpose programming language. It is used in web
development, data science, creating software prototypes, and so on. Fortunately for
beginners, Python has simple easy-to-use syntax. This makes Python an excellent
language to learn to program for beginners.
13. Project Dashboard Screen
Menu Bar
Tool Bar
Choose Button
for selecting the
input video file
Save Button
for save the
output file
Quit Button for
exit the software
14. Future Scope
In this project when we tested the video file we achieve 80% accuracy and in the
future that can be achieved by more than 80% by improving the generated audio
synchronization & audio recognition.
Future work will be devoted to better adapt machine translation to the style used in
dubbing and to improve the quality of prosodic alignment, by generating more accurate
sentence segmentation and by introducing more flexible synchronization.
In the upcoming days, we will make a device which help for those people who are a
tourist and that tourist are foreigner which native language is something else and tourist
place language is different and that device will live to translate automatically the visiting
place to tourist native language