kshitijdwivedi28/srfp_2021

Building Subtitles(.srt) And Transcript(.txt) file for a given Video using Speech Recognition, Wave Module, and Time Manipulation. Implementing forced alignment of audio and transcript based on word count, transcript, fuzzy search and duration of the video.

SRFP_2021

Speech Recognition and Forced Alignment of Transcript and Video

The project work aims to develop a Graphical User Interface (GUI) for building Subtitles(.srt) file for the input Video(.mp4).

Installation

1. Download the project through the below-mentioned link as a ZIP file from GitHub or clone it on your machine. Links - Project Link 1 or Project Link 2

2. Extract the Downloaded ZIP to a folder.

3. Download and Install Python on your platform through this link

4. Switch on to your favorite Integrated Development Environment (IDE) as the project is platform-independent (Windows / Linux / MacOSX) as well as IDE-independent (PyCharm, VS Code). I've preferred Visual Studio Code (VS Code) as IDE.

5. Open VS Code on your Platform and Navigate to File Button at the Top Corner.

6. Click Open Folder, browse to your extracted folder, and select that folder.

7. Navigate to the requirements.txt file from the left pane in the folder and install the mentioned modules. Use the package manager pip to install.

python pip install -r requirments.txt

8. Navigate to final_projec.py and run it. Equivalent terminal command -

python final_project.py

9. The GUI appears, for the project work requests for the input video file.
10. And as you click on the Generate subtitles button, you will have the resultant subtitles(.srt) file in the input directory of the video file.

Test Video Link

https://drive.google.com/drive/folders/18w6TaU7XvS1Gl9iuRyE8W76B6umFjiZy?usp=sharing

Demo Video Link

https://drive.google.com/file/d/1tmI8Xuoh8xbNd7Ceb-Qj1lXSgPCMi703/view?usp=sharing

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

Handcrafted by

Kshitij Dwivedi, B.Tech (Computer Engineering), Jamia Millia Islamia, New Delhi.

Project work completed. Waiting for Mentor's Approval.