Innovation is most definitely the fuel that drives technology. With an ever-growing market for new electronics, engineering adjusts according to the needs of the consumer. This requires innovative products that improve usability. LCD touch screens, faster processors and better camera capabilities have already become antiques. The next big thing is speech recognition!
Speech recognition (SR) is the translation of spoken words into text. Speech recognition is used in electronics such as mobile phones, tablets, computers and industrial electronics. Speech recognition aims to eradicate the need for typing and increase usability. Speech recognition in a broader meaning refers to Speech-To-Text (STT) Translation, Text-To-Speech (TTS) Translation with speech synthesis (producing machine generated voice), and voice recognition which involves recognizing the speaker among a group of speakers.
Figure 1: A Broadly defined SR System
Although Speech recognition software has been around for many years now, its accuracy is not enough for it to replace typing. Siri, Apple’s SR software lets you use your voice to send messages, schedule meetings, place phone calls and much more. Google and other leading electronic giants have their own SR software’s that allow you to do the same. Furthermore, Nuance Communications and many such companies produce SR software used in industries. Although these products have shown a pinch of hope for accurate SR devices, there is a lot to do!
SR falls under digital signal processing, a branch of electrical engineering. Although a speech recognition engineer is predominantly an electrical engineer, one has to be knowledgeable about many aspects of computer science, computer engineering, software engineering and also linguistics in some cases. An SR engineer applies knowledge of digital signal processing to create algorithms used in SR, to detect words spoken with uttermost accuracy. A SR engineer needs to be proficient in computer languages such as C++, C#, Java and scripting languages such as Perl, Python and even TCL (Tool Command Language) along with signals and systems, digital signal processing and analysis of algorithms.
The need for SR engineers is growing as Apple, Microsoft, Google, Nuance Communications, BBN Technologies and Yahoo expanding their SR sectors.
Many leading universities, including Florida Tech, now offer graduate level programs tailored for success in the SR industry. Dr. Veton Kepuska, Associate Professor of the ECE department at Florida Tech has remarkable research going on in the SR field. Florida Tech’s SR program is comprised of three core graduate level courses – speech processing, speech recognition and search and decoding in SR, along with many other required courses that fall under the systems and information processing domain of a M.S. in electrical engineering.
So, save your mobile phones with keyboards as a couple of years from now, they will be relics!