One of the most time-consuming elements of working as a journalist is the laborious process of transcribing content from audio or video files.
In the days before the internet, manual transcription was a workable option as deadlines were more structured than is the case today.
There has been a dramatic shift in the way people consume media content since the turn of the century, which has created a 24/7/365 news cycle.
Journalists are under increasing pressure to be the first to break news or features, making it difficult to keep pace in a landscape that has been dubbed ‘churnalism’.
The days when journalists could spend hours transcribing interviews are long gone, with media outlets desperate to get content into the public domain as quickly as possible.
As with many other processes in the modern era, technology has provided journalists with a solution to the problem of inefficient transcription processes.
By utilizing a transcription service online, journalists can slash the amount of time it takes to transcribe audio or video content.
The past few years has witnessed massive developments in artificial intelligence (AI) and machine learning (ML) that have transformed the way journalists work.
Read on as we look at how these emerging technologies are revolutionizing online transcription and why this is so important in this new era of multimedia.
Artificial Intelligence in Online Transcription
Automatic speech recognition (ASR) technology uses AI to interpret the sounds in audio and video files and converts them into text.
Manually transcribing this type of content can take hours, but AI can reduce the timescale to just a few minutes which boosts productivity.
Advancements in the technology have helped machine-generated transcripts deliver results that are as accurate as undertaking the work manually.
While AI is still a few percentage points short of perfection, it takes the bulk of the legwork out of the process to dramatically improve efficiency.
Some of the latest wave of online transcription services become smarter with each use, thus delivering greater accuracy each time they are used.
AI also negates the need for several people to be involved in the transcription process, giving journalists greater editorial control over their content.
However, AI in transcription works best when supported by human input, either from the journalist themselves or someone with proofreading skills.
The Advantages of Machine Learning in Transcription
We have briefly touched on some of the benefits of automating transcription processes, but there are undoubtedly many more we need to consider.
Significantly reducing the manual element of transcription is a clear advantage as this serves to ease the burden on the journalist.
There are also significant benefits to the media outlets given the excessive costs involved in paying someone to manually transcribe audio or video content.
Accurate transcription can be achieved in a much shorter amount of time, thus freeing up the journalist to work on more news or features.
Media outlets no longer need to have several people performing different tasks to create content, which helps them to reduce their overheads.
ML supported transcription also boosts security, which can be particularly useful if a journalist is working on a story with sensitive information.
Protecting sources plays an essential role in establishing and maintaining journalistic integrity, which makes the use of ML a no-brainer for any self-respecting exponent of the art.
AI & ML for Transcribing in Journalism
While some journalists still prefer to manually transcribe interviews, those who aspire to work at the top of their profession have embraced the modern way of working.
The New York Times’ national political reporter Astead W Herndon is a great example of this in action, with his smartphone a vital tool in his journalistic efforts.
He told the Columbia Journalism Review: “I record almost every phone call using an app which is attached to a transcription service.
“Even when I record I often take notes so that I know which sections of the tape to come back to, or if I’m working on deadline it’s a way to keep moving quickly.
“The app sends me a rough transcription of the phone call about five minutes after it’s completed.
“If I’m in-person I’ll use my phone as a recorder, and usually let the person know it’s on. I’ll also jot down notes and important times by hand.
“I don’t use a recorder for some person-on-the-street stuff, but in those instances I almost always go over the person’s quote with them, since they usually have less media literacy and I would not want a person to be surprised.”
Online Transcription in the Future
Although AI and ML are revolutionizing transcription processes, they have probably only scratched the surface in terms of where things will progress to.
ASR has been around since the 1950s, yet it is only in recent years where the technology has started to deliver what many experts have long been forecasting.
Generally speaking, most of the technology solutions in the sector fall short of delivering 100 percent accuracy, although some are closing in on that mark.
However, the technology’s ability to speed up transcription processes is undeniable and will only improve as it becomes increasingly smarter.
From a journalistic perspective, online transcription services will become part-and-parcel of how every media outlet operates in the next few years.
It will potentially make the industry more inclusive, opening the door for people with visual or hearing impairments to seek careers in the sector.
By establishing a platform for greater diversity in journalism, online transcription services may even help to restore confidence in an industry that often gets a bad rap.
About the Author/s
The New Jersey Digest is a new jersey magazine that has chronicled daily life in the Garden State for over 10 years.
