Can Speech Recognition Software Improve Transcriptionists’ Productivity?

So much has changed since the advent of speech recognition software in medical transcription. But speech recognition software, which is said to have made the lives of many a transcriptionist easier, is not without challenges. The technology, although widely available in many variants, falls short at times, requiring transcriptionists to QA and spot-check their copies for errors that may cost them millions in lawsuits and damages. The software is not intelligent (think Siri) as well—although not yet. Which raises the question: can speech recognition transcription really help with productivity? Can you really transcribe more dictations with speech recognition software? Absolutely, according to the results of a study titled “Speech Recognition as a Transcription Aid: A Randomized Comparison with Standard Transcription.” The study in question, which was published on the US National Library of Medicine website, provide empirical proof that speech recognition software indeed help transcriptionists to accomplish more dictation jobs. Researchers of the study divided participants into two groups—assigning one group to finish dictation jobs with standard transcription, and the other to use speech recognition software. Each group was then divided into subsets: endocrinology, full-time transcriptionists and full-time secretaries. Data results show that the use of speech recognition software has enabled transcriptionists to accomplish more jobs by 200 more, or up to twice as much as what those doing jobs manually have accomplished. The researchers of the study noted though they did not find any substantial benefit of using speech recognition in record keeping. However, the study’s authors also discovered that productivity seemed to be higher in certain subsets like psychiatry full-time transcriptionists and psychiatry secretaries. Some authors of the dictations—who were heavily accented or talk in a more conversational or casual manner—also tend to affect transcriptionists’ productivity. When removed from the subset analyses, secretarial and transcription productivity increased a bit, although not significantly. The study also said that the software was more beneficial to secretaries who work slower than others, and for transcribing longer recordings. A physician also expressed his satisfaction with the speech recognition software he uses in a letter sent to MT Daily. The physician, who wrote to MT Daily, said that that he barely had any problem with the dictation software he uses called NaturallySpeaking. In fact, he noted, he wrote the letter using the software. “At this point, I am 99 percent certain that I will be continuing to do my office progress notes using NaturallySpeaking,” he told MT Daily. While the physician pointed out some challenges in using the dictation device—especially when dictating punctuations—he didn’t discount the fact that the NaturallySpeaking software was highly accurate and relatively fast (50 words per minute). The physician also sees medical transcriptionist benefiting from the said software in upcoming years. As evidenced by the above accounts, medical transcriptionists, and anyone using speech recognition software for that matter, have to get the hang of using the software to eventually be able to do more with it. The manner of dictation also plays a role in increasing or limiting the productivity of secretaries and transcriptionists, which somehow points out that physicians, who are the dictations’ authors per se, should also be trained in dictation as well to make the lives of transcriptionists easier, too, so to speak. Hope this article has provided you some insight on the effectiveness of speech recognition software in medical transcription. How productive one can be may still vary depending on certain conditions, but data shows that in terms of initial output, speech recognition can surely help secretaries and transcriptionists get more jobs done.