5 Convenient Ways You Use ASR Technology Every Day

Automatic Speech Recognition, also known as ASR technology, is a type of AI technology that converts speech into written text.

Systems today can transcribe conversations in real-time with accuracy nearly on par with that of people, while also considering different languages and accents.

The History of ASR Technology

The 20th century made leaps and bounds in science and technology, and talking to machines, especially ones that could “understand,” was easy to imagine, but it didn’t seem likely for the near future. Yet people have always been driven by curiosity and a desire to make the impossible possible. This is especially relevant today with AI and its swift integration into everyday life. It makes you wonder what will be possible fifty years from now!

Research for ASR technology dates back as early as 1952, when Bell Labs (founded by its namesake Alexander Graham Bell, the inventor of the telephone) created “Audrey.” Audrey was a first-of-its-kind speech recognition system that could recognize numbers zero through nine. While innovative, the caveat here was that the system required its users to pause after saying each number. Ten years later, however, researchers were able to improve the system so that it could recognize standard words like “hello.”

Then, in the 1960s, IBM, the dominant force in the computer industry, developed the system, Shoebox. This computer could also recognize numbers zero through nine, as well as arithmetic commands like “plus” and “total.”

ASR technology soon caught the attention of DARPA (a U.S. research agency that develops and plans for technological surprises), and in the 1970s, they funded a program called Speech Understanding Research. Their research led to the creation of Harpy, which could recognize over 1,000 words by 1976.

The ball continued to roll faster, and the turning point in ASR technology arrived with the introduction of Hidden Markov Models (HMMs) in the mid-1980s. HMMs focus on the “statistics” of speech (for example, how likely one sound is to follow another). Consumer availability finally came in the 90s and resembles the ASR technology we use today.

Are We Using Machine Learning for Voice Recognition?

Yes, machine learning is being used for voice recognition. In fact, the term “machine learning” was coined around the same time as the initial research on ASR technology.

Once synonymous with the term “self-teaching computer,” machine learning enables ASR technology to learn patterns from audio data and put them into text.

5 Convenient Ways You Use ASR Technology Every Day

ASR has quietly become one of the most integrated technologies in daily life. Here are five ways that you use ASR technology every day.

1. Talking to Alexa, Siri, or Any Other Voice Assistant

Do you own a smartphone? Or have you at least used one? If the answer is yes, which it probably is, then it’s likely that you’ve used Siri, Alexa, or some other kind of voice assistant.

That means you’re using ASR technology.

Picture this. You’re alone on a Friday night, the house is empty, and suddenly a math question comes to you (as it would for anyone). “Hey Siri, what is zero divided by zero?”

The device ponders for a minute, and the silence is loud. “Imagine that you have zero cookies and you split them evenly among zero friends. How many cookies does each person get?”

The question is rhetorical, if not snarky. You feel embarrassed, maybe some shame, but the curiosity is killing you.

“See?” she continues. “It doesn’t make sense. And Cookie Monster is sad that there are no cookies, and you are sad that you have no friends.”

There was no reason to bring Cookie Monster into this. She could’ve just said “zero.”

In short, the device’s microphone records your audio and sends it to the cloud, where ASR algorithms analyze it and identify the basic sounds of human speech (i.e., phonemes). This information gets converted into text commands that prompt the device to complete the request, or unfairly tell you that “you have zero friends.”

2. While You're Driving

Newer infotainment systems in cars rely on ASR technology to help drivers stay focused and remain hands-free. This comes in handy (no pun intended) when you’re going 80 on the freeway, trying to finish your McDouble, and call your mom all at the same time.

Many newer cars integrate ASR technology by using virtual assistants, allowing for control over multiple functions like navigation, weather, and communication without someone ever needing to take their hands off the wheel.

When you use hands-free calling, dictate a destination to Google Maps, or ask your car’s infotainment system to play a song, ASR converts your spoken words into digital commands that the system can understand and act on.

3. Reading Closed Captions

Closed captions are everywhere: YouTube, Netflix, livestreams. And they’re not just for people who are deaf or hard of hearing, but for anyone who is in a noisy environment.

When you can’t hear the video you’re watching over the sound of your chips, what do you do? You turn on closed captions. That’s ASR technology. The system listens to the dialogue, converts it into text in real-time, and then syncs the words with the video you’re watching.

Beyond just entertainment, closed captions help during online meetings, lectures, and virtual conferences, allowing listeners to follow along without missing key points. ASR technology has turned what used to require hours of human transcription into a feature that works behind the scenes instantly.

4. Translating a Foreign Language

Have you ever used Google Translate to talk with someone who doesn’t speak the same language as you? ASR technology is powering this, too. When you or the other person speaks into your phone, the software transcribes your speech into text using ASR. Then, it converts that text into another language.

The great thing about this is that if you want to go to a café in Paris but don’t know French, ASR technology will have your back. Without ASR doing all of the heavy lifting upfront, live translation apps wouldn’t understand you at all.

5. On the Phone with Customer Service

We’ve all heard the classic: “Please say or press 1 for billing.”

ASR technology is at it again. Interactive Voice Response (IVR) systems use ASR to recognize what you’re saying and send you to the right department. It analyzes your voice in real time and passes that information along to the system.

Now, instead of listening through a dozen options, you can just say “billing” or “technical support,” and ASR technology will take care of the rest.

And when it works properly, it’s much faster and saves you from furiously pressing 1 over and over.

Final Thoughts on ASR Technology

From hands-free driving to voice assistants, captions, and translations, ASR technology has become something we regularly depend on day-to-day. What once seemed like an idea out of science fiction is now an everyday convenience.

—

Danielle Mundy is the Content Marketing Specialist for Tier 3 Technology. She graduated magna cum laude from Iowa State University, where she worked on the English Department magazine and social media. She creates engaging multichannel marketing content—from social media posts to white papers.