How to use voice recognition and how does it work?
By Indeed Editorial Team
Updated 23 November 2022
Published 10 May 2022
The Indeed Editorial Team comprises a diverse and talented team of writers, researchers and subject matter experts equipped with Indeed's data and insights to deliver useful tips to help guide your career journey.
Voice or speech recognition is the term given to software that can listen to and analyse a piece of speech before performing the command spoken. Today, voice recognition software is available on phones, smart speakers, laptops and even home appliances. Businesses and individuals are learning to utilise voice recognition to streamline and simplify daily tasks. In this article, we explain how to use voice recognition, examine what it is and explain when it's used.
What is voice recognition?
Voice recognition is becoming increasingly popular both at home and at work. It can fill a number of niches, such as controlling smart homes, commanding phones and tablets and instructing smart speakers. The device operating voice recognition often requires a trigger word, such as 'Hello' or 'OK'. When spoken, the device knows that it now has to listen to the following command. The device then analyses the speech so it can act on the command given.
How to use voice recognition
To get the most out of this type of software, it's essential to understand how to use voice recognition. Different businesses and individuals use a range of different features of voice recognition based on their needs and preferences. There are several ways you can use voice recognition as a tool, including:
1. To dictate and save text
Voice recognition is useful to have on a laptop or computer as it may help employees write up their notes faster. For example, those with a laptop powered by Windows can access this feature by opening up 'Speech Recognition' found within 'All Programs' and then 'Ease of Access'. Once they've elected to either press the microphone button or say 'Start listening', the laptop picks up and transcribes everything said into a document. This is valuable for doctors and nurses as it means they can spend more time with patients and less time typing up notes.
2. To print and send documents
A common command that employees use voice recognition for is sending and printing documents. You can perform this task on Windows-supported devices by asking Cortana, 'What can I say?'. Cortana then talks the individual through which phrase they're required to say so that the device understands what it has to print. This is usually 'print'/‘send' followed by the printer or sender's name. An employee also makes sure they're on the page they wish to print or send before making the command. This guarantees that Cortana can't misinterpret which document to send or print.
3. To write and send emails
Another command that employees typically use the voice recognition feature on a Windows-operated device for is writing and sending emails. This process is similar to asking Cortana to dictate a piece of text. The user first opens their email app and clicks the 'Compose a new email' button. Then they ask her to 'Start listening'. In a clear voice, the employee recites their message. After checking for any errors, they're able to ask Cortana to 'Fill fields' by providing the contact details for the recipient. Finally, they ask Cortana to send the email.
4. To browse the web and complete forms
Another function employees commonly use voice recognition software for within the workplace is to browse the web. It's often useful to ask Cortana to research a question while working on something else. To do this, the employee tells Cortana to ‘Start listening' before they ask their question. After a few seconds, she relays the answer and provides a list of related Internet pages. When using Cortana to fill out a form, whether online or in a Word document, it's important to use the 'Fill fields' command to ensure they format the information correctly.
How does voice recognition work?
Two main things happen before voice recognition works. The first is the measurement of a voice's biometrics. This includes the accent of a voice and how fast the speaker is talking. To make the request manageable, a device segments every word spoken into smaller parts. This means that the device isolates each word, dividing it into different tones. Once the device has done this, taking only a few seconds, it shares what it's understood from the command with the speaker.
The second thing required to happen before voice recognition can take place is the enacting of a command. For this, the spoken command is sent to the device in the same way as a button being pressed. The action then subsequently happens. When a new device is first used, this process lags a little, but as time goes on, the device's thinking time speeds up as it becomes used to a specific voice, in addition to abbreviations and acronyms. Some devices do this by digitising and storing each command for future reference or by using neural networks.
When is voice recognition used?
Voice recognition is being increasingly used in commercial environments. There are several different occupational contexts where voice recognition is commonly used. Some of these contexts include:
Various businesses use voice recognition to improve how they offer their customer services. For example, an employee can serve a customer much quicker, with more customers served in a day overall when they use this type of AI technology. This is because when a customer contacts a company to discuss the problem, the device picks up the type of query they have and forwards them to the correct professional. The alternative is a customer waiting in large queues to find out that they're getting forwarded to another department, leading to customer dissatisfaction.
Speech recognition systems are also useful in several settings, whether the user is a journalist recording facts for a story or a solicitor working on a case. With ease, they simply ask their laptop to record the meeting. This comes as opposed to the alternative, which is to try to keep up with a discussion by scribbling down a series of notes. It's rather impractical to do this because key details are sometimes missed or illegible. This causes a great deal of avoidable stress. Note that it's important to get the consent of those being recorded beforehand.
The versatility of modern-day computers allows them to perform multiple tasks at once in such an environment. For example, the user can use their laptop to project a presentation that's key to the meeting while simultaneously recording it. The user can also automatically upload a copy of the recording to the cloud in case something happens to their device. The overall benefit for both an individual and the company they work for is that there's no added expense of buying separate devices for recording and uploading.
To give patients common discharge information
Patient discharges happen every day. The discharging process requires a nurse to attend to each patient and give them certain information about aftercare and what the next administrative steps are. In most cases, this is repeatable to more than one patient. Therefore, to help make the nurse's job easier, some hospitals and care homes deploy AI devices to perform this task for them. Their interactive nature means that a patient can ask the system any questions they may have in the same way that they'd ask a human nurse.
This allows nurses to focus on other situations where the use of an AI is impossible or less appropriate. It also means that the discharging process for outgoing patients is less rushed.
Advantages of voice recognition
There are several advantages associated with voice recognition:
Productivity: Voice recognition can help boost productivity. For example, in a healthcare setting, doctors can dictate their notes, and secretaries can organise patient administration tasks much quicker when booking GP appointments.
Accessible: Voice recognition is much more accessible than using a mouse or a keypad to make a command, as a user only has to speak aloud within appropriate listening distance. This makes it ideal for users with more specific physical needs.
Grammar: If a user is unsure of how to spell a sentence or ensure that it's grammatically correct, they can speak the sentence aloud to the speech recognition system during a speech to text command. This inserts the correct version of the text into a document.
Please note that none of the companies mentioned in this article are affiliated with Indeed.
Explore more articles
- What is lead nurturing? (Plus how it works and benefits)
- 12 reputation management software to build brand presence
- How to scan multiple pages into one PDF using two methods
- What is the psychology of sales? (With impacts and examples)
- What is trade protectionism? A definitive guide (with tips)
- The purpose of leader pricing and how it applies to you
- 8 online public speaking classes and why they're important
- A guide to content ideas: definition, examples and tips
- What is alt text and why is it important for businesses?
- Product-market fit: how to assess it and tips to achieve it
- Understanding consulting graduate schemes (with salary)
- Examples of digital strategy (with components and tips)