The Korean Phonetic Aligner automatically segments a speech file into component words and phones. It accepts as input a speech file (.wav) and its corresponding Hangul transcription in UTF-8 encoding (.txt) and returns a Praat TextGrid file. A UTF-8 encoded transcript file can be prepared with Notepad (or 메모장) on Windows or TextEdit on Mac OS X. The aligner follows the romanization convention summarized here.
  1. Fill out the form below with valid information.
  2. Click 'browse' to upload the desired Korean Hangul transcription file (.txt) and the desired wave audio file (.wav) from your computer.
  3. Click 'Submit' to have your request processed. The result should be ready for download.

First Name *
Last Name *
Email *
Affiliation *
Upload transcription file (*.txt) *
Upload Wav file (*.wav) *
I authorize the collection of the information provided in the form.
Sample Alignmenti) example.txt ii)example.wav

The front end of the Korean Phonetic Aligner was written by Tae-Jin Yoon at Cheongju University, Korea. The interface between Praat and HTK is adapted from the Penn Phonetics Lab Forced Aligner. The website was developed at the University of Toronto Scarborough as part of the project “A Corpus-based study of Korean dialects: microvariation and language universals” funded by the SSHRC Partnership Development Grant (Principal Investigator: Yoonjung Kang).
Tae-Jin Yoon & Yoonjung Kang. 2013. The Korean Phonetic Aligner Program Suite.
Questions? Comments?