Voice Recognition Set to Grow in Healthcare

In a recent Healthcare IT News article, they wrote about a KLAS research study that found that the speech recognition market is ready to grow. Here’s a small portion of the article:

Providers report a demonstrable return on their speech recognition dollars, according to a new report from KLAS. Participants of the study indicated benefits of speech recognition such as staff reductions, improved report turnaround times and increased physician satisfaction.

“The speech recognition market is ripe for healthy growth,” said Ben Brown, author of the report. “Currently, less than one in four hospitals use the technology, however, in light of meaningful use and the benefits providers point out in this study, we expect it will assume a more prominent place in the role of clinical documentation.”

It seems like a bit of journalistic spin to say that speech recognition provides a “demonstrable return.” My personal experience tells me that users either love or hate speech recognition. The article does aptly state that it requires some up front investment to learn voice recognition and access the long term benefits that voice recognition provides.

The other obvious part of the report is that Dragon still dominates the voice recognition landscape. I recently also got an email from Eric Fishman of EMR Consultant, EHR Scope, EHR TV, etc fame (and also an advertiser on this site) about a new voice recognition, dictation and transcription software they’re distributing called Frisbee.

They have a bunch of videos showing Frisbee transcription software in action on EHR TV. I found the one called Frisbee, Dragon Medical and EMR Workflow pretty interesting.

I could see this type of software providing the platform for the future of the transcriptionist. Neil Versel recently posted the news that the Medical Transcription Industry Association (MTIA) will be changing their name to the Clinical Documentation Industry Association. No doubt transcription companies are looking at ways to survive. One of those ways will be for the transcriptionist to go beyond just transcribing to assisting with the clinical documentation (including the complicated ICD-10). Seems like Frisbee’s voice recognition into the EMR with the Frisbee routing capabilities for doctors approval and sign off could be an interesting workflow.

I’m not quite as bullish on voice recognition as the report linked above, but there’s no doubt that voice recognition will continue to play a role in healthcare. Especially as it continues to improve its recognition ability and becomes integrated with mobile devices.

About the author

John Lynn

John Lynn

John Lynn is the Founder of HealthcareScene.com, a network of leading Healthcare IT resources. The flagship blog, Healthcare IT Today, contains over 13,000 articles with over half of the articles written by John. These EMR and Healthcare IT related articles have been viewed over 20 million times.

John manages Healthcare IT Central, the leading career Health IT job board. He also organizes the first of its kind conference and community focused on healthcare marketing, Healthcare and IT Marketing Conference, and a healthcare IT conference, EXPO.health, focused on practical healthcare IT innovation. John is an advisor to multiple healthcare IT companies. John is highly involved in social media, and in addition to his blogs can be found on Twitter: @techguy.


  • Demonstrable? I guess that means good?

    As mentioned, voice recognition tends to be a love/hate relationship.

    I love the idea of it, but hate the implementation.
    When I use it for “simple” items, I find myself taking more time than if I just typed from the beginning.

    I’d also bet most docs who uses it for transcription would still rather speak into a recorder, send it off for transcription while the doc does something more desirable.

    Additionally, is your desire as a patient to see your doc come into the room with one of those headsets on? (yes, I’m aware of the “mr. microphone” capability.)

  • I think it means easily demonstrated, but either way it translates to good.

    The interesting part of the Frisbee video link above is that they are using a recorder which sends it off to transcription which puts it into the EMR. That’s why I found the workflow interesting.

  • @John … to your “love or hate” perspective the piece on the KLAS study said:

    “Providers indicated there is room for improvement. They would like speech systems to learn more quickly. Some customers report accent challenges with speakers who are not native English speakers. Some also report that adopting a front-end speech technology system tends initially to be disruptive to workflow.

    Though there is a learning curve, and some providers report a workflow interruption while adopting speech recognition technology, physicians are getting used to it and embracing it, the report shows.

    “Generally the KLAS survey finds that a culture change occurs as providers begin to fall in love with speech-recognition technology,” said Brown. “Once physicians realize that by spending a few more minutes up front they save time and money in the long run, many are sold on it.”

    That last comment doesn’t sell me on whether KLAS quantitatively defined the delta between hate and love. Just saying that physicians will embrace technology that has a higher front end if the payoff is there at the backend … when the value of the result is not quantified … hasn’t panned out anywhere else so why with voice recognition either? (I think this is the official definition of hype).

    There’s a way to go yet and until clear quantitative (and qualitative clinical) value leaps out front loaded energy will not get expended.

  • I have used Dragon for years. Every version gets better but just when I think they have got it right I discover with continued use that it is still not quite there yet.

    I am puzzled by the enterprise solutions that use transcriptionist proof reading to back up the speech recognition. How is that cost effective over regular transcription?

    My ideal speech recognition device would be a form factor like the Philips SpeechMike with the speech engine inside the microphone. After you train it you can carry it with you, plug it into any USB port and the computer thinks its a keyboard.

Click here to post a comment