You must have heard about google support, Microsoft Cortana, apple, but do you know that all this technology works on the speech recognition system and voice recognition system. Now you would think that this is what is a voice/speech recognition system and how it works. Today, we are going to tell you this about the people, how this technology works and what is its history.
What is Speech Recognition System?
It is a mix of computer software and hardware that has the ability to explain the human voice and answer it accordingly.
Using this technology does not require any mouse or keyboard to operate your mobile or computer. You can only command your mobile or computer by voice. If you have to open an app then you can also open it using the voice command like if you want to open Facebook, you will open it to facebook when you say it open facebook.
More About Voice Recognition Technology
In 1952, a scientist named Audrey created a voice/speech recognition system in the bell laboratory but it used to work only on digits.
In 1962, the company named shoebox redeveloped it, and now it also started explaining some of the English words. The company again developed it so that it got the ability to understand the English 9 consonant and Vowel.
When the U.S. Defense Department saw it in 1971, he started his own research program on Speech Recognition Technology. Carnegie Mellon developed software called Harpy, which could easily understand 1011 Word. He also searched for a system that could easily understand the logical sentence. But their biggest drawback was that all of this used to be on a speech pattern.
In the 1980s, a speech recognition system, called the Markov model, used digital data to explain the sound. And in 1987, with the help of this technology, a doll named Julie was made, which used to talk to children, but it was a flaw that one word had to take some break after speaking.
In the 1990s a company named dragon created software called Dragon Dictate, it was the world’s first Voice / Speech Recognition Software, which the common man could also use. In 1997 the company made further improvements and named it Dragon Naturally Speaking, it could understand 100 words.
In the 2000s, Google created the Google Voice Search Application for the iPhone, which responded to the statements spoken by humans by matching the store data in the data centre. In 2010 Google created a Personalized Recognition application for Android, and through it started recording voice queries. And it has created a database of about 230 billion words. And even Google Assistant has launched its own software, which you can use in your android mobile.
Seeing Google, Apple also did a delay less personal assistant named Siri, whose name is Siri. Looking at the success of these two companies, Microsoft also developed a software development called Cortana.
How Speech Recognition System works?
All of you will know that when we speak something, a wave is produced. Mobile or computer converts those waves into digital signals through ADC translators.
The ADC Translator splits the computer or mobile analog signal into small samples. And convert them into digital formats. After that filter by filtering the noises removes that noise. After that, normalizing the sound leads to the constant level. Because the way of speaking every speech is different.
After this the sound is divided into small parts like the thousandth of a second. And finally these small parts are matched with the store digital signals in the system. As if you said GO, then this system will process it after match with the store data in memory or match with G signal G, and match with O signal O. And our computer immediately shows us the output GO.