Taken from Tribune India.
The Indian Institute of Technology (IIT) here has developed a software to convert written text into speech.Pulkit Sharma, a PhD scholar, said the software, Text-To-Speech (TTS) systems, would be a revolutionary step for visually challenged people. It would enable them to use computers efficiently, as it would pronounce every word they would write, he said.“The system has numerous other applications. It can be used as a desktop assistant and can read out SMSs in mobile phone,” he said, adding that the system could also be used in public announcement systems, to read directions and destination in a GPS device.Pulkit, who hails from Hamirpur district, has been working on the project under the guidance of Dr Anil Kumar Sao, Chairperson, School of Computing and Electrical Engineering, IIT-Mandi.Others helping him out in the project include Sarvesh, Nivedita, Priyanka and Manav.Dr Sao said the work on TTS systems was being carried out by a consortium of 12 institutes headed by Prof Hema A Murthy of IIT-Madras.The other institutes of the group include CDAC-Mumbai, CDAC-Thiruvananthapuram, CDAC-Kolkata, IIIT-Hyderabad, IISC-Banglore, IIT-Guwahati, SSNCE-Chennai, PESIT-Bangalore, DA-IICT-Gujarat, IIT-Kharagpur and IIT-Mandi.Dr Sao said each institute would build a TTS system for one Indian language using a common framework and the IIT-Mandi was working on developing a TTS system for Rajasthani language.“Although the synthesised voice using the current TTS systems is quite intelligible, but research work is being carried out to make the voices more natural and human-like,” he said.Dr Sao said it was a unique project where researchers from 12 different Indian institutes had come together to develop TTS systems for 12 different Indian languages.It would largely help the visually challenged people, he said, adding: “If a visually challenged person wants to read news, the software will easily read it out if it is installed in his device.”He said the project, which is sponsored by the Department of Electronics and Information Technology, was given to the institute in 2012 and the Centre for Development of Advanced Computing (C-DAC)-Pune was helping them out in evaluating the system.Pulkit said: “The TTS systems are classified into two categories: Unit Selection Based (USS) and Statistical Parametric Based (SPS). Voice synthesised using USS systems is more natural, but such systems require more memory to be stored thus cannot be used in devices with less memory (e.g. mobile phones).”IIT-Mandi was working on to reduce the memory required to store USS systems thus enabling the device with low memory to use them, he said.The TTS system would be handed over to the C-DAC, Pune, by the end of the year and the centre would make it available in the market at a very nominal price, he added.
Source: Tribune India.
No comments:
Post a Comment