Welcome to the home of the Aligner.
Aligner is an automatic phoneme segmentation system for a number of languages. It currently supports English, French, German and Italian.
With the help of speech recognition technology, it can detect the start and end times of words and speech sounds (phonemes) within recorded utterances. It requires an orthographic transcription (typically a text file) on what is spoken in order to do so. Aligner is typically used by phoneticians, linguists and engineers that deal with spoken language research. As it is easy to use, it can be useful also for other fields such as media editing.
Aligner is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
Aligner is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You will need to install HTK (available from http://htk.eng.cam.ac.uk) in order to make use of Aligner. HTK is a leading Speech Recognition Toolkit utilizing Hdden Markov Models (HMM).
The latest version of the Aligner is 2.4beta, available from the following link: Aligner-2-4beta.tgz.
The version is supporting English, French, German and Italian. Support for Polish is on the way. See the chapter on contributing for how you can help extending Aligner's coverage.
See the file INSTALL for installation instructions.
See file USAGE on how to use the Aligner.
See file BUGS for bugs and known issues.
See file AUTHORS for contributing authors and a short development history.
See file LICENSE for the full GPL V2 license text
The long term goal of Aligner is to support sufficient languages to cover all the phonemes of the international phonetic alphabet. Then, it will be easy to retrain Aligner for any of the world's languages, and help researchers to also study minority and endangered languages efficiently and with the same kind of tools that are available for the well-researched languages for years.
You can contribute by extending the Aligner to further languages, donate resources such as speech corpora or already trained HMM files or processing components that can be included in or added to the Aligner such as pronunciation lexica or automatic grapheme to phoneme conversions. You are also welcome to help with the various tasks in improving the code of the Aligner itself.
Whether you want to simply use or actively develop Aligner, it is recommended to subscribe to the aligner-user mailing list. Technical discussions are addressed on the aligner-developer mailing list.
Both lists are unmoderated lists for registered users so you must register on the mailing list page to participate. It is encouraged to briefly introduce yourself when you give your first question, bug report, recommendation or other contribution.
If you notice a bug or have a question, please use the aligner-user mailing list. You need to register to the list before, so you can also receive any answers from the community.
If you can not modify Aligner to your needs for some reason, you can ask for help on a voluntary or paid basis. As Aligner is free software, basically every sufficiently skilled person can help you with extending Aligner for your language or purpose. You may also ask the people at the aligner-user or aligner-developer list. Please note that the terms of the license are binding and make sure that they are met.
Finally, you can purchase professional support or custom developments on a for hire basis from the original author of the software at email@example.com.