Below are the projects that I am currently maintaining.
Script Analyzer is an implementation of the Script Analysis Framework proposed in my PhD dissertation. It enables extracting various quantitative metrics quantifying scribal behavior by analyzing them in terms of their constituent strokes.
It also allows the reconstruction and visualization of the trajectory a.k.a ductus, along with the capability to view the different types of strokes.
(Click on the thumbnail to play the screengrab)
The source code can be found in Github.
Konkanverter : Konkani Statistical Machine Transliteration Engine
Konkanverter is a finite state transducer based transliteration engine that performs statistical machine transliteration between three different scripts (Devanagari, Kannada & Roman) used to write the Konkani language. The statistical machine transliteration system consists of cascading finite state transducers combining both rule-based and statistical approaches.
You can access the project from here.
Avalokitam : Tamil Prosody Analyzer
Avalokitam is an open source prosody analyzer for the Tamil Language. It constructs a parse tree of the input text based on the prosodic syllable patterns. Based on this parse tree, the input verses are analyzed for all the six basic elements of Tamil prosody: eḻuttu (Letter), acai (Metreme), cīr (Metrical Foot), taḷai (Linkage), aṭi (Metrical Line) & toṭai (Ornamenation). The meter of the verse is then recognized by matching the elements with the elaborate and complex rules of Tamil prosody.
Anunaadam : Tamil to IPA Transcription Tool
Anunaadam is an open source transcription tool for Tamil. It transcribes the given Tamil text and produce the output in IPA characters. It attempts to aid the pronunciation of Tamil even to non-native speakers of the language. Apart from IPA, the tool can also transcribe the language in several other Indian scripts.
Aksharamukha : Transliteration Tool
Aksharamukha aims to provide script conversion between various scripts within the Indic cultural sphere. These include historic scripts, contemporary Brahmi-derived/inspired scripts, scripts invented for minority Indian languages, scripts that have co-existed with Indic scripts (like Avestan) or linguistically related scripts like Old Persian.
Apart from the simple mapping of characters, Askharamukha also attempts to implement various script/language-specific orthographic conventions (where known) such as vowel lengths, gemination and nasalization. It also provides several customization options to fine-tune and get the desired orthography.
Aksharamukha as of now supports 71 scripts and 6 romanization methods.
Adinatha – Tamil Brahmi Font
Adinatha is a Unicode font for the ancient Tamil Brahmi script. In fact, it is the only font that is designed to display Unicode Tamil-Brahmi properly. Adinatha is based on the existing Brahmi encoding and largely draws upon Dr Iravatham Mahadevan’s Early Tamil Epigraphy for the shapes of the glyphs. The font contains both OpenType & Graphite table and hence properly renders in a wide variety of systems.
You can read more about the project from here.