Sci & Tech

Persian Voice Recognition Software Developed

Persian Voice Recognition Software Developed
Persian Voice Recognition Software Developed

Iran has launched its first-ever online voice recognition software in Persian language named Nevisa Live.

The new software also allows business and home users to spell the Persian words in a Microsoft word document, TechRasa reported.

Nevisa’s latest version has doubled the vocabulary to 120,000 words.

“You can upload an audio file recorded in advance and the software has the ability to transcribe the audio,” said Hossein Sameti, a software developer and professor at the Faculty of Software Engineering at Sharif University of Technology.

“Another feature of the website is the ability to recognize the accent and the tone of the speaker and to make a personal profile of each user.”

Various versions of the software are available to meet the needs of different users. These include a home edition for personal use; an advanced edition for writers, publishers, students and teachers; a professional edition for organizations, businesses and typing centers; a legal edition for lawyers, jurors and official facilities; and a medical edition for preparing radiology, MRI and CT reports.

Nevisa benefits from a noise robustness technique that helps the software better understand the user’s voice even in noisy environments.

The software is developed by Asre Gooyesh Pardaz Company, a knowledge-based company specialized in artificial intelligence and signal processing.

The team of developers includes experts and graduates from Sharif University of Technology.

The trial version of the app is available on the company’s website for free.

In the international arena, Dragon Naturally Speaking still leads with software that works with a host of major spoken languages, including Spanish, English and French.

The developers of Nevisa could likely use their technology and even sell it to a large company like Dragon, as the latter currently does not support the Persian language.