At first, refer to the given tables to get the characters you want. After a time this will become automatic.
Try live at https://www.eludev.fr/bangla
The application is written with Elm, which allows for:
- easy and clean development
- very small application size (33 kb for the generated javascript)
- easy communication with javascript
- simple, consistent and efficient tooling
We tried to use a logical transcription close to Bangla script logic:
We chose to stick with the classical names of vowels, which do not correspond to present day bangla phonetics : a, i, u, e, o, r', with their 'long' variants a', i', u', e' and o'.
The choice between the two sets of vowels ("full" and "matra" vowels) is made automatically based on the context.
We use _ prefix only in _( to get a regular ( since we use ( for candrabindu ँ
We use two suffixes:
- ' mainly for aspirated consonants (k' g' c' j' T' D' p' b' to get খ ঘ ছ ঝ ঠ ঢ থ ধ ফ ভ), but also for ঞ as n', শ as s', anusvara ং as (' and visarga ः as :', and for the 'long' vowels (see above)
- and " used for ষ : s"
To use ' or " as quotation marks, insert a space before them.
The basic consonant forms include the 'inherent vowel' a. Bangla script uses ligatures when there is no vowel between two consonants. For example: nd is translated as নদ , which also stands for nada To represent the nd sound as in "andar", you must mark a virama (transcribed as '-', the minus sign) between n and d: n-d is translated as ন্দ, as in an-dr : অন্দর
Any character not recognized is rendered unmodified.
MIT