This article was updated in April 2023.
Jot this down: Transcription services aren’t just beneficial for those of us who can’t write legible meeting notes. They’re also key to the future of marketing. With the artificial intelligence (AI) behind transcription software improving by the day—thanks to major advances in automatic speech recognition—there’s little reason to miss the boat.
Believe the Hype
How can AI transcription transform your marketing efforts? Let’s count the ways:
- Captioning videos. Video captioning has always been important for inclusivity but that’s not the only reason to consider it. According to Publicis Media, 80% of those who use captions don’t have a hearing impairment. That’s because we’re watching videos everywhere now—in public, at work, next to a sleeping baby—places that make listening to video problematic.
- Turning speeches, video tutorials and webinars into marketing collateral. Oral presentations take time and effort to produce. Why not repurpose that content into things like social media posts, downloadable reports, blog posts, email campaigns and any number of other marketing materials?
- Enhancing SEO. Once you have transcripts from the above recordings, you can use them to help boost SEO for those videos by providing content for search engines to crawl and rank. Add links in relevant spots throughout the transcript and you can boost SEO even more.
- Building your help center. Transcribe customer onboarding/troubleshooting calls and use them as the foundation for a searchable online knowledge base. You can use them to start building your help desk chatbot scripts as well, since those should be conversational and engaging anyway.
- Transcribing subject matter expert interviews. As a writer, this is the area in which I have the most experience. I’m happy to report that a good transcription app can seriously reduce the time it takes to turn phone interviews into articles, blog posts and white papers.
- Transcribing market research. Market research is critical for initiatives like product launches, and AI transcription increases its usefulness exponentially. Once you have primary research like phone surveys and focus groups transcribed, you can do cool stuff like analyze the conversations to determine sentiment and measure the frequency of certain words associated with a product or idea.
- Clearing the air. Marketing teams operate on an increasingly global scale and virtual meetings are becoming the norm. But even when teams across geographies speak the same language, unique dialects (there are 160 different English dialects, for example), poor connections and other factors can muddle communication. Transcribing these meetings can help set the record straight.
Choose Your Poison
That’s just a handful of uses; there are many more. But listen up. While transcription software is advanced, it’s still not perfect. There are dozens of companies offering AI transcription services, and what makes one better than another is very subjective.
It’s likely the best service for video captioning won’t be the best for in-person or phone recordings. Plus, some offer auto-generated scripts only while others offer human-assisted transcription as paid add-ons.
No matter what—even if you pay top dollar—you’ll need to review the transcript and do some editing yourself. As Rev, a leading software provider in the space, points out, even 99% accuracy equals about 15 errors every 1,500 words.
Start by comparing a few free versions or plugins and see if they do the job. You can also experiment with what might already be available on your computer. Google, Microsoft and Apple each offer speech-to-text functions that work well for direct dictation, and social platforms like Facebook and YouTube provide limited captioning capabilities. Recorded audio and phone conversations, however, are trickier and require more specialized transcription apps (PC Magazine reviews several of them).
Even many of the paid transcription software tools offer free versions or free trial options, so experiment liberally. The number of errors you’ll get as you experiment will vary widely; only you can decide what error rate is acceptable for what you need.
Just don’t make a choice and be done with it.
AI—and automatic speech recognition specifically—is big business and innovation is happening rapidly. Keep an eye on those advances to utilize the best (and most cost-effective) solutions. Make sure to stay ahead of the curve when it comes to turning audio into pure marketing magic.