The VACHANTAR-Rajbhasha is a Speech-to-Text translation system. The two major technologies used under this system are Speech Recognition & Machine Translation. The Speech Recognition technology is used to convert the input speech into Text and the Machine Translation Technology is used to translate Source Language to Target Language. The system takes English Speech as an input and translates it into Hindi text in the domains of Administrative, Finance, Agriculture, Small-Scale Industry, Information Technology, Health Care, Education and Banking.
The system combines a Speech Recognition Engine and MANTRA-Rajbhasha, a machine translation system. The Speech Recognition Engine takes English speech as input through a microphone and generates a typed English text as output. This output becomes the input to the Translation Engine that translates it to Hindi.
- Interactive Graphical User Interface.
- Live and pre-recorded speech as inputs.
- Translation of the document at runtime using voice command.
- Document alignment and editing using voice command.
- Supported input file formats: .txt, .doc, .rtf and .wav.
- Source document format retention in the output.
- Interactive preprocessing tools for phrase-marking, date and proper noun identification. Word addition, grammar creation and updating facility.
- Modes of Translation: Direct, Interactive, Step by Step and Selective Mode.
- User friendly selection of multiple outputs.
- In-built thesaurus for selection of contextually appropriate synonyms.
Centre for Development of Advanced Computing
Applied AI Group, 4th Floor,
C-DAC Innovation Park,
Pune - 411 008.
Phone No.: (020) 25503314 / 15