2017fandrei / ForcedAlignment

Graphical utility for forced alignment using aeneas, an interactive audio player

ForcedAlignment

ForcedAlignment is a graphical tool to automatically synchronize audio and text.

Contact: [e-mail]

Goal

ForcedAlignment generates a html file with an audio player and a list, from a text file and an audio file. The text and audio file must match. The text is highlighted as the speech progresses. The text is split by delimiters like .(full stop) !(exclamation mark) ?(question mark) ,(comma) ;(semicolon) in units for easy following. By clicking on such an unit the reading begins at that moment. A text file and an audio file exists as samples. The generated temp_csv_file.csv and temp_txt_file.txt can be deleted.

System Requirements, Supported Platforms and Installation

System Requirements

1. Python 2.x or 3.x or later
2. Aeneas https://github.com/readbeyond/aeneas/
3. A html5 compatible browser (for the audio tag)
4. Depending on the browser the audio file can have many audio formats.
5. Python must be in the PATH
6. Tkinter and other python modules

Supported Platforms

Linux, Mac OS X, Windows

Installation

Requires the dependencies.

Usage

Select the text file, the audio file, and the language for both files and a html file and run the script.

License

ForcedAlignment is released under the terms of the GNU General Public License Version 3. See the LICENSE file for details.

Files

ForcedAlignment_python2_1.py for python2

ForcedAlignment_python3_1.py for python3