Logo Lanfrica

jouml/NCHLT-Praat-Plugin

Domain:

natural language processing

Record type:

software
Creator:
jou
Host:
Praat plug-in for exploration and acoustic analysis of the NCHLT speech corpora of South African languages # The Praat NCHLT plugin moved to Codeberg Version 1.1 of the plugin (July 2025) supports the latest SADiLaR folder naming pattern and is available on Codeberg. **This repository is obsolete!** ## Description This plugin enables Praat to search in the orthographic transcriptions of the NCHLT Speech Corpus and open the audio files of corresponding search results. The NCHLT Speech Corpus contains orthographically transcribed broadband speech corpora for all of South Africa’s eleven official languages. Each language corpus consists of a large training suite (trn) and a smaller test suite (tst). The NCHLT Speech Corpus must be installed on your machine (at least one language) before using this plugin. The corpus may be obtained from SADiLaR. After launching from the Praat *Open* menu, you first select one of your installed languages, pick one of the suites, and specify a search pattern. The XML transcription is loaded into Praat and searched for the pattern. You can further refine the search using speaker attributes (gender, age, location). When you are done with searching you can view the results in a table (including orthographic transcription, speaker ID, age, gender, and location) and open corresponding audio files for acoustic analysis. ## Manual A detailed manual is included. ## Recquirements - Praat 5.4.x or newer - NCHLT Speech Corpus (at least one language) ## Installation Instructions available on Codeberg. ## Usage When launched the first time, the plugin asks for the base directory of the NCHLT Speech Corpus. This is the directory/folder that contains the language-specific sub-directories called `nchlt_ ` (old format) or `nchlt.speech.corpus. ` (current format) (` ` is a placeholder for the 3-letter language code): ```x base directory │ ├──nchlt_ │ ├── audio │ │ ├── │ │ │ ├── nchlt_ _ _ .wav │ │ ... │ └── transcriptions │ ├── nchlt_ .trn.xml │ └── nchlt_ .tst.xml │ ├──nchlt.speech.corpus. │ ├──... │ ... ... ``` After se …