Before diving into the complexities of our industry’s current state, I’d like to take a moment to reflect on my last post, ‘Speaking of AI: Why I’m Confident in the Future of Professional Transcribers.’ While that conviction remains, this article aims to provide a more nuanced view of the challenges and opportunities we’re facing as of late 2023.
Technology is shaking the foundations of real-time transcription, an industry long reliant on human expertise. While advancements like Automated Speech Recognition (ASR) have brought incredible speed and some level of accuracy, they still fall short in areas that humans excel at, such as meaning-for-meaning translations.
The Roots of Real-Time Transcription
Historically, services like CART and TypeWell have led the field in real-time transcription. These aren’t just software; they’re services powered by skilled human transcribers. Their unique styles meet a broad range of accessibility needs, from education to the corporate world.
AI’s Current Limitations
While AI performs well in CART-style verbatim transcription, it faces hurdles in meaning-for-meaning translation. Errors in real-time speech recognition can snowball when the same text is further paraphrased, leading to what’s known as ‘compounded errors.’
Workforce Impact
Our recent survey shows that over 40% of TypeWell transcribers felt their current semester’s workload didn’t meet their expectations, suggesting some normalization post-pandemic. Technology isn’t the only factor driving these changes; educational environments, pay rates, and user expectations also play a part.
The Complexity of Job Satisfaction
Switching to roles like editing AI-generated transcripts may sound straightforward, but it may not offer the same level of job satisfaction. Transcribers find fulfillment in the mastery of their craft, something that merely correcting AI mistakes might not provide.
An Uneven Future
We’re at a turning point. AI-driven services are improving rapidly, making the landscape more complex. The decision to use humans or machines for transcription isn’t just about costs or convenience; it’s a complex narrative involving legal considerations and user needs.
This is a summary of Kate’s LinkedIn article published 8-31-2023: https://www.linkedin.com/pulse/shifting-landscape-real-time-transcription-what-weve-been-kate-ervin