High-Tech communication: voice and emotion recognition

"Communication among the human being is dominated by spoken language, therefore it is natural for people to expect speech interfaces with computer: computer which can speak and recognize speech in native language"

Hello everybody!

We communicated to each other since ages: 18 000 years ago our ancestors painted their daily life stories on walls in caves, then invented clay tablet, papyrus, parchment and finally- paper. Nowadays we digitalized even this last one – and switched from paper printing to ebooks with a touchscreen. In general, we can communicate in 3 ways: by using signs- letters/drawing, body gestures- for example hand moves, and sound – our voice.

When the first computer was invented the only way for communicating was by using the keyboard, then was invented “clicking” mouse… Together with technical development, we got a new way of communicating with our personal device – by touchpad and screen touch. The latest invention is to use gesture and voice recognition. Today I would like to invite you to see how does it work when is really helpful and how can it be adapted to my interactive touchscreen in a neighborhood of Drents Dorp!

As I was aware by my roommate who is working in a care facility, some of older people have a problem with body movement, and it could be hard for them to click on interactive device… So what should I do then… I realized that my smartphone has a tool which recognizes my voice and gestures: and as the last one are not available, I decided to take a look deeper into a voice recognition system.

In general whole system works as follow: 

• voice (sound waves) is detected by microphone
• electrical impulses generated by microphone is transferred to a sound card where acoustic signal I being converted to digital one
• software which is a hearth of our system converts digital signal to phonemes, and then words
• depending on a device we have, text can be displayed on a screen or our “command” can be recognized and processed as required (going to menu, turn a volume up or any other)

It seems to be simple- in reality, it requires a lot of effort of engineers from different areas to combine all together in order to have high quality, “intelligent” software.

This system is really helpful for people with gesture/movement disabilities and the one who is blind or can barely see. Those kinds of people can be found in any place and should not be excluded in intervention as mine. In general, this software can be used for simple commands like “next, previous, main page, close, open etc.” It also can be used together with speaking software: those who have a problem with reading would be able to hear it.

Of course, there are some cons: it is hard to recognize jargon and kind of training session should be arranged to teach how we should communicate with such device – but I guess, that it is not so problematic when we compare it to pros which we can get:

- possibility to recognize different languages

- no isolation, available for people with disabilities

- access to information and news

- better socialization within a neighborhood

And voice recognition is just a first step in High-Tech communication: let’s imagine emotion recognition…

Cheers,
Natalia

References:

Corea, F. (2017). Artificial Intelligence and Speech Recognition for Chatbots [online] KD nuggets. Available at: https://www.kdnuggets.com/2017/01/artificial-intelligence-speech-recognition-chatbots-primer.html [Accessed 29 May 2018]

Agrawal, M., Raikwar, T. (2016). Speech Recognition Using Signal Processing Techniques. International Journal of Engineering and Innovative Technology, 5(8), 65-68. Available at: http://www.ijeit.com/Vol%205/Issue%208/IJEIT1412201602_13.pdf [Accessed 29 May 2018]

Chun-Feng, L. (2003). Understanding the CMU Sphinx Speech Recognition System [online] Research Gate. Available at: https://www.researchgate.net/publication/268414151_Understanding_the_CMU_Sphinx_Speech_Recognition_System [Accessed 29 May 2018]