This paper discusses a spoken query system developed for accessing some household electronic devices. The developed system enables the user to access these devices remotely by calling the system using a GSM modem/ Mobile phone. The spoken query system has mainly two modules i.e. interactive voice response (IVR) and automatic speech recognition (ASR) modules, which can be developed using open source resources. In the development of ASR models, the system specific data are collected from people of different ages and genders. The issues of data preparation, Training and Testing are performed by using HTK as a speech recognition tool. The system can recognize sample data i.e. the voice commands independent of vocabulary size, noise, and speaker characteristics. The recognized voice command as a text file will be used as an input into the microcontroller that is responsible to control the electronic home appliances. This speaker independent interactive voice response system is based on HMM (Hidden Markov Model) which is found to give a relative improvement of accuracy by more than 50%.

IVR, Speech, ISR, HTK, Training, Testing, HMM

[1] Armando Roy Delgado, Rich Picking and Vic Grout, “Remote-Controlled Home Automation Systems with
Different Network Technologies”, Centre for Applied Internet Research (CAIR), University of Wales, NEWI, Wrexham, UK
[2] L. R. Rabiner, “Applications of Speech Recognition in the Area of Telecommunications,”IEEE Workshop on Automatic Speech Recognition and Understanding Proceedings, pp.501–510, 1997.
[3] S. Goel and M. Bhattacharya, “Speech based dialog query system over asterisk pbx server,” in 2nd International Conference on Signal Processing Signal Processing Systems (ICSPS), Dalian, Jul 2010.
[4] Asterisk open source communications, http://www.asterisk.org/home

[5] Muhammad Ali Mazidi, Janice Gillispie Mazidi, "The 8051 Microcontroller and Embedded Systems Using Assembly and C “, Second Edition.
[6] S. R. Suralkar, Amol C.Wani, Prabhakar V. Mhadse “Speech Recognized Automation System Using Speaker Identification through Wireless Communication”, IOSR Journal of Electronics and Communication Engineering (IOSR-JECE), e-ISSN: 2278-2834,p- ISSN:2278-8735.Volume 6, Issue 1 (May. - Jun. 2013), PP 11-18
[7] Nurul Fadzilah Hasan, Mohd Ruzaimi Mat Rejab and Nurul Hidayah, “Implementation Of Speech Recognition Home Control System Using Arduino”, ARPN Journal of Engineering and Applied Sciences, VOL. 10, NO. 23, DECEMBER 2015 ISSN 1819- 6608
.
Author’s Profile


Dr. Manoj Kumar Deka, received the M.Sc. degree in Instrumentation in the year 2007 and Ph.D. degree in Signal Processing in the year 2011 from Gauhati University, Assam. He achieved INSPIRE Fellowship sponsored by Department of Science & Technology (DST), Govt. of India, New Delhi during his Ph.D. work. Presently he is working as Assistant Professor in the Department of Computer Science & Technology, Bodoland University, Assam. His area of interest includes Signal Processing, Embedded system, Artificial Intelligence.