Speech Recognition is also perceived as automatic speech recognition (ASR) and voice recognition comprehends the verbal phrases and terms and renovates them to a machine-readable configuration. By restoring spoken audio into text, speech recognition technology lets users control digital appliances by talking rather than employing conventional tools such as keystrokes, buttons, keyboards etc.
This voice recognition technology aims to make our daily life easier. It helps people to interact with the appliance without touching that very appliance. Amazon’s Alexa, Apple’s Siri, etc are remarkable voice recognition software. It will create a great revolution in the recent future in the field of technology.
Contents
History of Voice Recognition Technology
Voice recognition technology stands in today’s situation through a gradual modification that goes back to the 20th century to those early days.
Let’s explore the history and progress of this technology step by step.
FIRST STAGE
Audrey is the first system using voice recognition technology, concentrated on numbers, in 1952 which was designed by Bell Laboratories. Audrey could analyze 97% of its command.
After ten years, IBM’s innovation was the second advancement in this sector.
They invented the machine named “Shoebox” like a calculator.
This device could realize 16 words and ten digits 0 to 9.
SECOND STAGE
In 1971, the US Department of Defense and DARPA had run a project named Speech Understanding Research (SUR). And this marked the massive research initiatives in the field of voice recognition. The ultimate result of this program was Carnegie Mellon’s “HARPY” voice recognition system. HARRY could interpret more than Thousands of words.
THIRD STAGE
Hidden Markov Model (HMM) is the rebellious creation in the field of voice recognition. It worked based on the statistical method. This HMM could discern ten thousand phrases.
Using this “Julie” a doll, was the awe at that juncture. Julie could even reply to an individual. Uncommon voices being words could be identified with this technology.
FOURTH STAGE
The personal computer was available for the first time in the world in the 90S. With this advancement, Dragon Dicta made an upheaval in voice recognition technology.
Now the question is what is Dragon Dicta?
Dragon Dicta is the first marketable voice recognition commodity that could understand the biological speech of up to hundreds of words per minute.
FIFTH STAGE
Google Voice Search with 80% accuracy was the first modern advancement. With this chance, google data center turned on collecting words from searchers. And in this way, google voice search could foretell what someone is certainly mumbling.
FINAL STAGE
Fin 2011 Apple constructed Siri. Actually which is the up-gradation of google’s voice search technology. Siri also worked based on the cloud. It collected billions of words from the searcher and stored them in the cloud.
Eventually, in the next, it matched the words with the formerly stored words.
It could also recognize the unknown sounds being a word.
Amazon’s Alexa is another innovation in the field of voice recognition.
If you are looking for the origin of Alexa you have to look back to 2014 when Amazon first announced it. The idea was first used and inspired from the beginning of “Star Trek: The Original Series” where the “computer voice” and conversational system” were used.
Now in this final stage technology world is blessed with much-advanced speech recognition software. All these are highly scrutinized. Different technology companies have launched different speech recognition software with different features.
Like Microsoft has launched Cortana, Apple has launched Siri. And the like software are Zubtitle, Happyscribe, Dolbey, Verbio, Sonix, Google docs, Speech write digital, Otter, Braina, e-speaking, etc.
How Does Voice Technology Work?
Voice and speech recognition technology has grown rapidly in the last few years. Now people depend on technology on a large scale. Various research papers of great researchers from different parts of the world show that most people nowadays spend 3 to 4 hours on their electronic devices.
Truly speaking most people are supposed to use electronic devices for their daily needs. In a time of using these types of electronic devices, they have to use different types of technologies like software.
Speech recognition software is one the top used software these days. There is so much speech recognition software like Siri, Cortana, Amazon home, etc.
The accuracy level of these types of software is also praiseworthy.
Many of these can do their job up to 95% accurately. Research says 50% of all web searches have been conducted using voice search technology in 2020.
And it is valued that at the end of 2024 about 30 billion people will use this technology on their devices.
Do you know how it works?
OK before comprehending this, just know the basic theory. As the computer is not the human brain so it faces some difficulties in recognizing the voice. So first we shall discuss the difficulties faced by the computer or like devices in recognizing the voice.
OUTCRY
Humans can grab the effective fractions of the speech from the noises and background hues and cry. But a computer can’t differentiate between these useful and unnecessary words and catch that as a whole.
The quickness of linguistic transmission
people with a biological brain can recognize rapidly spoken speech, high and low tuned voices meshed with sentiments and expression. Most of the speech recognition technology face problem in catching words more than two hundred words per minute.
Accent
Humans are used to accents of their own. They even don’t understand all of the accents of different parts of the nation. So it is not expected that a computer can understand all the dialects of the different parts of the world. This is why speech recognition technology faces problems in recognizing words.
So in a word, it can be said that human speeches are full of enthusiasm and manifestation but the computer is an electronic device without any emotions that’s why this speech recognition technology faces problems. And 100% accuracy has not appeared.
Now we shall discuss the much-expected question of how speech recognition technology works?
Voice recognition means understanding human speech with the help of software by a computer or electronic device. A computer converts the human voice into text with the assistance of a microphone and speech recognition technology.
The mechanism of this technology is demonstrated below
Speech to text transformation
For completing this process a computer has to maintain some sophisticated steps. People create vibration while speaking. The analog to digital converter (ADC)
Analyze this analog wave into digital data by seizing several gauges. The computer digitizes the sounds for removing unexpected disturbance and make various bands of frequency. Voice recognition technology temples different sounds at a certain speed in their cloud. So when a computer heard a sound, firstly it adjusts the speed of that sound to match with the sound already stored in their cloud.
Next, the computer divides the signal into minor components as short as a few hundredths of a second to match this segment with the known phonemes.
Then the program matches the phonemes with the previously stored words in their cloud. And this a very difficult task to be done through a statistical model.
Breaking of waves
Breaking of waves is the most important task for voice recognition technology. Because the accuracy of any specific voice recognition software depends on this task. Ventured waves are as minor as 1/16000th of a second. Computer collects bits through this procedure that can easily be understood by a computer.
Recurrent Neural Network (RNN)
The human brain is the wonderful creation of Allah and it can store a word in its memory for a lifetime and can match the word with the word previously known to it. But a computer can’t act like this. So the scientists tried to develop something which can store the words and next time can match with that. So they developed a set of algorithms that are eligible for storing a massive set of data and processing that to convey as the diagram given to it as production. This is called a neural network as it tries to work as a neuron of the human brain.
Neuron Networks have verified to be extremely productive by correlating thick learning to comprehend patterns in photos texts and speech.
Recurrent Neural Networks can influence its future outcomes. When it reads a letter it also predicts the next letter. By giving an instance this can be made more clearer.
Suppose sitting before your computer you have just whispered OXY it is highly probable that it will say GEN next to it. And this RNN recoups the earlier prediction for its future predictions.
Traditional neural networks are not used nowadays because they can’t save their previous prediction in memory that’s why they do mistakes in future predictions. This why RNN has made a coup in the history of speech recognition technology.
Long Short Term Memory (LSTM)
Long Short Term Memory (LSTM) is the legacy of RNN. Recurrent Neural Networks can’t filter extended lines. To overcome this scientists invented this new technology.
This is structurally different from Recurrent Neural Networks. As we know that Recurrent Neural networks consist of one configuration.
But the Long Short Term Memory (LSTM) consists of 4 structures. Like cell state and 3 others. Cell state allows data to flow through it. There are three types of gates. They are called input gate, forget gate, and output gate. Cell state is conserved with the subsidy of these three gates.
And with the aid of gates, any input can be strengthened or removed. The sigmoid is another feather of this Long Short Term Memory that runs two functions. Whether they will pass data that was inputted or on data will be passed.
All this about the technology of speech recognition.
There are varieties of speech recognition software. They are different from each other based on the working method.
Varieties of voice recognition software
Speaker-dependent software
Speaker-dependent networks are handled by the individual who will operate the systems. These systems can recognize the command and words with an accuracy of 95%. They initially comprehend and evaluate the characteristics of their user’s voices because they are greatly conditional on their speakers. But once they are used to their speaker’s voice they can be treated as very efficient speech recognition software. This is very familiar among the individual.
Speaker independent voice recognition software
Speaker-independent voice recognition system is widely used for public uses. This system has to respond to a wide range of words. Because in public uses speakers can’t be fixed. This system is motivated to concede to a word regardless of who is speaking. The efficiency of this system is expectedly lower than the speaker-dependent system
Command and control voice recognition software
this system is used to guide and control tools employing voice commands. Jobs like commencing the programs, browsing through websites, and other processes can be effortlessly achieved. This system is run to a remote place through wifi.
Discrete input voice recognition software
This system ensures the accuracy of word identification. Sometimes it can identify words up to a Hundred percent accurately. But it requires a pause after each phrase. A discrete input voice recognition system has a default grammar with keywords and non-keywords. This innovation interests a discrete input voice recognition system with grammar.
Ballooned grammar stays effective until the operator gives rise to a new choice. This network advantageously offers a flexible proportion between minor length default grammar and enormous functionality facilitated by huge size grammars to thereby mitigate the possibility of false and erroneous recognition.
Continuous input voice recognition software
You will find a lot of software for recognizing the voice. A very few software provides continuous voice input recognition facilities. The software building company designed the software for the continuous word flow and analyzing those words too.
The main difference you will find from other voice recognition software is that they evaluate fewer words and provide the best results within moments. Continuous input voice recognition software is mostly used in medicine and healthcare sectors.
You will find some software like Apple Dictation (a free app for Apple devices). Dragon Medical One is one of the most popular voice recognition software used in healthcare which can transfer voices into text while a doctor is something in a tested microphone.
Natural speech input voice recognition software
This type of software is more advanced than any other type of software. Because this type of software can understand fluent and speedy speaking. It can appreciate more than 160 words per minute. Nowadays, in all modern devices.
Automatic Speech Recognition (ASR) System In Home
Technology has reached a mythical position these days. Your home can be furnished thoroughly with technical assistance like speech recognition. You are thinking how, right now? Ok, let’s make you clear.
Every home-appliance can be controlled without touching that very appliance. It’s maybe fictitious to you. But it’s true.
Door
Let’s start from your door.
Back from your workplace, you are standing in front of your home, and you want to open your door without touching your door. This is possible. Just buy a speech recognition system and set it on your door inputting your voice with a specific sentence like ” please open the door” to the system. Then the door will open automatically.
Fan
Now you are in the compartment and the weather is hot. Feeling the heated need to switch your fan on.
Just say “switch on the fan” the ASR will start automatically without touching the switch button.
Light
Now think about the situation in the night you fell asleep and at midnight you wake up and need to switch on the light, just say “on the light”. Light will be on automatically.
CCTV
You can control the CCTV screen with the assistance of ASR. At night suddenly you hear that your baby is being cried in the nursery. Just say on the screen.
Your monitor will be on the nursery’s camera on your screen.
TV
Sitting before the TV screen you need not use the remote. Simply command the system on your tv. Your television will be on automatically. You are boring to a channel and want to change that?
Let out “change the channel”.
Daily Routine
For maintaining your daily routine you can take the assistance of your speech recognition system.
Waking in the morning asks the system for your Schedule.
The system will tell you about the schedule for that day. When will the conference start? What is the schedule next to your conference that will let you know?
Calling and Receiving
Your system can receive the call automatically with your command.
Someone is calling you?
Command your ASR system to receive. The system will receive the call and connect you.
Kitchen
A Voice recognition system is used in the kitchen also nowadays. You can give a command to your Instant Pot for warming water.
Like many others, the appliance is also available.
For all these home purposes Amazon’s Alexa can be your best buddy.
Automatic Speech Recognition (ASR) System in CARs
In a voice recognizing car, on steering a button is settled for receiving a command from the driver. It is used to regulate features like satellite navigation, playing music, call and receiving voice calls, etc via a Bluetooth connection.
Think about the circumstances. You are in the driving seat and you don’t know the exact address of your destination. You can ask your voice recognition system about the exact place of destination. Your voice recognition system will show you the path to your destination on your screen.
Again think of another catastrophe.
You from a long drive, are a little bit tired and looking for a coffee shop. And as you are on an unknown highway that’s why you don’t know any coffee shop in that very locality, in this condition you can take the assistance of your voice recognition system.
Ask your voice recognition system for the address of a coffee shop and the system will show you the address and route using satellite.
Traffic Signal
Whether you are on a known way or an unknown way you don’t know the traffic signal in that very direction. So at the time of your journey, you can check the traffic signal with the help of a voice recognition system.
The system will search the signal ahead of you. And the system will suggest an alternative way to your destination.
Controlling Radio and Media
On driving, you can feel bored sometimes. You can take refreshments by listening to music.
If your car is fitted with a voice recognition system, command the system for playing your favorite music of yours. The system will search the music for you.
Again if you want to know about the current news while driving you can command your system for news channels. The system will search the channel automatically.
Calling and Receiving Voice Calls
On driving, you may need to call someone or receive voice calls from someone.
You can do this with your voice command not touching the button of the phone.
Suppose you are on the way to your meeting. And you are late for some reasons.
You should let your boss know about your delay. Ask your voice recognition system to dial the number of your boss.
The system automatically will search the number from your saved contacts and dial.
Again someone can call you, in that case, you can command your ASR system to pick the call.
Controlling Over Speed
Over speeding is one of the major causes of accidents. Sometimes the drivers lose their control on speed.
In statistics, it is indicated that about 60-70% of accidents are caused by over speeding. So if the over speeding could be checked, the accident rate will be curtailed.
In this case, a voice recognition system can play a vital role. If it is connected with the speed meter via Bluetooth it will give an alarm when a car will cross a programmed speed limit.
Suppose the driver sets a limit of 100 kilometers per hour in his system.
Next time when he will cross the limit the system will give a signal automatically. Like “Don’t cross this limit, it is very risky” or in the language anyone sets.
Reading above, what are you thinking about this technology? Should you use this technology in your car?
The answer is yes obviously. Because this voice recognition system in cars helps you a lot from different directions. It is safe. It will help you to find your destination if you get lost any time on your way.
It will give you a chance of being refreshed on your driving time. Just with a command, your favorite song will be added to your playlist. So all of you should set this technology on your cars for a safe journey.
Automatic Speech Recognition (ASR) system in Healthcare
The speech recognition system has taken a revolutionary change in the sector of healthcare. It is used in the different departments of healthcare.
We will discuss the use of voice recognition in healthcare in different steps.
Telemedicine
Telemedicine is the phrase that denotes the treatment by a doctor over the phone.
In an emergency, a patient can’t go to the doctor or hospital instantly. That’s why he has to take treatment from home with the assistance of a telephone or mobile phone.
Especially in a pandemic situation, when people can’t go out then this is very useful.
But the problem is sometimes, the doctor can’t catch the words spoken by the patient appropriately. Then if the speech of the patient can be transmitted into text it would be easier for a doctor to identify the problem.
To get rid of this problem voice recognition systems can be used.
Because in that case, the speech of the patient will be transmitted into text before a doctor.
Restoring the history of the patient
The history of the patient can be restored in the cloud with the help of a voice recognition system.
Next time when that patient will come to a doctor, then the history of that patient can be rechecked by the doctor within a few seconds.
And this very first process than traditional typing.
Dragon medical is used on a high scale in this sector.
Medical Checkup
Now in many medical checkups, a voice recognition system is used on a high scale.
The heartbeat of a patient can be heard by the voice recognition system.
And in this, the accuracy level can be checked with the newly invented voice recognition system used for healthcare.
Using of Voice Recognition System in Advanced level
Science is crossing its level every day. Today which is out of imagination tomorrow that will be in reality.
Using the voice recognition system in Medicare is going towards a position that may be like imaginary thinking in Today’s world. Scientists are working on the project that, by analyzing the voice of a person some kinds of diseases can be identified.
Unambiguous Prescription
It is often found that the handwriting of doctors is not readable by the general public.
So the medicine is not used properly. Sometimes people get the wrong medicine from the ambiguity of the handwriting of the doctors.
It also happens that the patient goes to bed condition from taking the wrong medicine.
But the use of voice recognition can stop this.
When a doctor suggests any medicine for a specific patient it will be transmitted into text automatically.
Automatic Speech Recognition (ASR) In Lift
People with a disability mean a human being with a problem in any of its organs.
Some may not have hands or someone may not speak. But in this competitive world, every man having or not having disabilities has to run in everyday life.
Suppose a man without hands works in an office. And he has to come to the office using the lift.
What will happen to him? He has to depend on others for his everyday communication.
But if there is an elevator controlled by voice?
It would be easier for that person with a disability. Now the question is what is an elevator controlled by voice?
Simple.
This is a kind of elevator that can be controlled by voice using voice recognition technology.
The Voice recognition system converts the language of the user into the apparent language of the computer.
But this system in an elevator can recognize a limited number of words. If the user speaks words that are not taught to the elevator previously, it can’t recognize that word.
But this type of voice recognition works more accurately.
And this accuracy is needed in the elevator. Otherwise, an Embarrassing situation will be created. Suppose getting into an elevator you have given the command to go to the 8th floor. But the elevator has taken you to the 9th floor. So the accuracy level is the first choice in the elevator. It uses Easy Voice Recognition technology for training purposes.
Arduino Uno is another technology that is used for controlling systems.
Now the question is what is Arduino Uno?
It is a microcontroller board found on the datasheet (ATmega328P).
And it is formulated by Arduino.cc which is an open-source electronics outlet. It consists of a manual programmable circuit cabinet and software. It is evolving outstanding with the people nowadays. The system can be equipped with a maximum number of thirty-two words.
After all, the use of voice recognition is increasing day by day. Because it is the demand of time. People wish to do all the things technically in the present era.
Benefits of Automatic Speech Recognition System
Faster
People can’t write as fast as they speak. Normally it takes more time to write than speak. But the speech recognition system is that which can transmit the words into the text as fast as speaking. That’s why people give preference to the Automatic Speech Recognition system for handwriting. Because the ordinary policy of handwriting can downshift the communication method.
Accuracy
The accuracy level of the Automatic Speech Recognition System is now out of doubt. There are so many companies that offer different types of voice recognition software. Specifically used for a home office. Some software is used for elevator cars etc. But most of the software offers high-level accuracy. In fact in the present world, 95-98% (percent) accurate voice recognition software is also available.
Hands-free Work
Voice recognition offers hands-free work. As we know the use of Amazon’s Alexa in the home. Nowadays people can switch on their light fan on other house appliances without touching the switch button. This hands-free work is also offered in other sectors. Now doctors can give prescriptions without touching their pen. People can run the elevator without touching a button.
In the business sector, this hands-free work has taken a revolution. Voice recognition system offers effortless note-taking, proficient eye communication. These types of installations would like imagination without the use of a speech recognition system. Think about another situation, you are on a drive and your hands are busy with steering. And you need to talk with your friend. In this situation, a speech recognition system will help you.
Speaking With Robots
Speaking with robots!!! Reading this sentence you might be astonished that what is needed to talk with the robot!
The Actual scenario is, the use of the robot in many sectors is increasing day by day. Like the preliminary interview is taken by the robot in many institutions. Once a role is performed by a human being that can be performed by a robot the next time.
The interview is conversational. At the time of the interview, the interviewee speaks, and the interviewer listens. So if you want to make a robot as an interviewer the robot must have the quality of understanding about the language of the interviewee. At this time a speech recognition system is needed. As the robot has to understand the language of the interviewee and transmit it into text.
Controlling digital devices
Suppose you want to control your digital devices without touching them. Then a speech recognition system is a must. Because if you, without touching your devices, want to control your devices then something is needed which can catch your command without your touching.
And that “something” is the speech recognition system. Now an example can be given for your better understanding. Think about your Alexa. What does Alexa do for you? Alexa conveys your command to your devices. When you just say your tv to start. Alexa conveys this command to the tv. With the help of software, tv is on then.
Listening without reading
“Listening without reading” seems an awkward matter for you for the first. Because you are a healthy person without any disabilities by the grace of Allah. Now guess about the person who has some disabilities like can’t look at the screen for a long time. But he wants to read the newspaper or other necessary documents. In this case, a speech recognition system is the best option for that kind of person. A speech recognition system will transmit the words into a voice for his listeners. Then the listeners will be able to enjoy the news without reading.
Criminal Investigation
Nowadays, offenders do some criminal activities over the telephone. Sometimes criminals threaten the general people using the telephone. Because they feel or think that if they use the telephone or mobile phone in criminal activities like threatening they can’t be identified. But the Speech Recognition System can analyze the voice of the criminal and in this they the criminals can be identified. Or think about another aspect, some criminal-minded persons are planning to do a crime talking over the phone. And the recording of the group call is at the hands of the police. In this situation, police can find the offenders using a voice recognition system as this system can analyze the voice of a person.
Call Center Operations
Today’s world is full of competition. From people to the establishment all are busy with their self-development. Many institutions have call center service. They provide instant assistance to their customers. All the companies have some special or regular customers. And they are privileged than others. As in call center service, the face of the customer is not shown. The call center managers can’t detect their special customers. In this circumstance, voice recognition is the best option for companies. Using a voice recognition system they can differentiate their special customers from the general.
Nowadays, a voice recognition system is used to a great extent in the healthcare sector. Because there are so many advantages of using a voice recognition system especially in healthcare.
Decreases wasted time
Clinicians can avoid wasting time by using a voice recognition system. A patient enters into a doctor’s chamber firstly the doctor listens to the problems of the patient. Then he makes a prescription. Here is the use of a voice recognition system. If he uses a voice recognition system then the system will transmit his voice into text and the patient will take the printed prescription.
Savings of Money
In a hospital, there are so many posts of typewriters. Their only job is typing text. In this sector, if the hospital authority uses the voice recognition system, they will be able to save their money. As in that time, fewer typewriters shall be needed. Because a voice recognition system can type faster than a typewriter.
Clinician satisfaction
Doctors feel relaxed if they get some time for recreation. Normally doctors pass a very busy day with patients and documentation. This documentation can be run with the assistance of a Speech Recognition System.
Disadvantages of Speech Recognition System
There is nothing in this universe that is full of advantages. Everything has some disadvantages. Speech Recognition Technology is not the exception. It also has some disadvantages. Now we will go through the disadvantages of the speech recognition system.
Imperfect Accuracy
The speech recognition system is not working with 100% accuracy. Sometimes it catches misleading commands from the user. We know accuracy is paramount in the healthcare sector. In this sector, every single word is important as it is related to the treatment of human beings. Suppose after detecting the problems of a patient a doctor suggests a tablet of 1 mg for that very patient but the voice recognition system transmits that as 1g. This will create huge trouble in the body of that patient.
Knottiness of jargon and phrasing
The speech recognition system can’t differentiate between the speech of the user and the jargon of the circumstance. Very often it translates the jargon and writes some text at its own will. The phrasing is another problem. In every sector, there are some technical words of their own. If those technical words are not used in Standard English then the voice recognition system can’t grab those words. In that situation institutions have faced great trouble with the speech recognition system.
Multiple Users at a time
In different parts of the world, people use different accents though they speak the same language. Even in a country, people from different areas speak in different tones. The problem lies here. Every Voice recognition system company uses a standard accent in developing their software. The system doesn’t work If a user speaks in a different accent from that standard accent. Again some people speak quickly. If the user speaks quickly the system will transmit the voice incorrectly as it requires slow and clear speaking. Furthermore, if two or more persons speak jointly at a time the voice recognition will work perfectly.
Time Cost and productivity
It seems that using the voice recognition system is time time-consuming. But it is not true all the time. Because the voice recognition system requires nursing from time to time. It especially requires training. This training session is very time costly and it needs an expert which costs money. The productivity is not also satisfactory all the time. Very often it causes grammatical errors. Again as it requires slow and clear speaking, so the user has to speak slowly which is not time-consuming moreover it is time costly.
Bodily Side Effects
Excessive technical Use has some effects on the human body. The Voice recognition system is not different at all. If a person uses this system regularly he will face some problems. It causes hoarseness, dry mouth, muscle fatigue, temporary loss of vocal strain, etc. As in the time of using voice recognition systems, people don’t talk naturally that’s why they face these types of physical problems. So if a person wants to use this technology regularly he must have learned the way how he can protect himself from these types of bodily side effects.
Limited Vocabulary
A speech recognition system can save a limited number of words. When a user uses words out of the saved words the system shall not work in the circumstances. So this is one of the marked disadvantages of voice recognition systems.
Training
Training is another disadvantage of the speech recognition system. Because it requires an expert person who can train. It requires huge time and money.
All these are the Disadvantages of voice recognition technology. Though the advantages of voice recognition systems are greater than the disadvantages.
Our View Regarding Voice Recognition Technology
Straight speaking, we are. Very optimistic about the future of this growing technology” voice recognition technology”. Once it was out of our imagination that a device can understand our command more surprisingly it can also translate them into text. It can carry command to any appliance which is voice recognized.
It was out of our imagination that without touching a device like a fan, television, the door can be opened. But now it is true. So from this point of view, we are very optimistic. As we already have discussed some disadvantages of the voice recognition system. Like, still now 100% accuracy is not achieved. There may be a 2-5% misinterpretation in this system.
We hope that in the future this voice recognition system will work 100% accurately. Furthermore, there are also some disadvantages like it can’t differentiate between the speech of the user and the background noise. We hope in future this technology will achieve this target point. Moreover, it will be able to catch the speech from the different users at a time. As of now, it can’t catch the command we two or more persons speak jointly.
We are hopeful about the other issue. That is insufficient vocabulary. Now the voice recognition system can’t catch the words out of the saved words in its cloud. But we are hopeful that in the future it will enrich its vocabulary so that it will be able to catch every word spoken by its users. It is not imaginary thinking.
All this is outward. But we also have some visionary thinking regarding voice recognition technology. What does it work today? It can catch the command from the user and it can transmit the voice into text. But our visionary thinking is that one day this voice recognition technology will drive to a place like, it will be able to give a reasonable reply also. What is a reasonable reply?
Suppose when you ask your friend a question. He doesn’t give the gist answer only. He throws some relate-able words also. This is called a reply. After all, it can be said that in the future talking between a human being and an electronic device will be like the conversation between two human beings. And its usage in the different sectors will increase day after day.