Olympus Dictation With Dragon Voice Recognition
If you’ve been using the Olympus® Dictation Management System thus far, you know how effective it is at managing your dictations, supporting playback and recording functions, and powering your delivery to a transcriptionist.
Your transcriptionist is certainly aware of the magnificent abilities of the software. With a foot pedal in place, she can easily listen to files and type up documents—as well as receive, manage, and file dictations effectively.
What neither of you may be aware of is that you are doing it the hard way. Allow me to explain.
The Dragon® speech recognition system, available in a number of editions, can be leveraged in such a way that:
- You create dictations on your handheld device
- Upload them to your computer automatically
- Have them run through speech recognition
- Deliver them to your transcriptionist(s) for final correction and document production
Instead of having your transcriptionist type out an entire dictation, she only has to edit the document produced by Dragon, and then put that in its final form.
In this post, we’re going to discuss some of the ways that you can successfully integrate speech recognition into your dictation-transcription workflow.
In order for you to be able to pair the Olympus Dictation Management System (ODMS) with Dragon NaturallySpeaking, you’re going to need to meet a few requirements.
First of all, we’re going to be discussing the functionality available within R6 of ODMS. While DSS Player Pro R5 can use Dragon, it doesn’t work in quite the same way. There’s no speech recognition queue to process jobs automatically, and the document is not attached to the dictation file.
If you just got the Olympus software and need some help setting it up, here are three excellent resources to consult:
- Highlights from the latest version, R6.4
- Setup tips for authors
- Setup tips for transcriptionists
If you’d like to know more about the Olympus Dictation Management System in general, see here.
Now, concerning Dragon speech recognition, you need to have one of the professional-grade offerings in order to work in the Olympus environment. That means one of these currently-available editions (or earlier supported versions):
- Dragon NaturallySpeaking Professional 13
- Dragon NaturallySpeaking Legal 13
- Dragon Medical Practice Edition 2
There’s also a little setup work to be done in Dragon. Pay attention to “Configuring Dragon for Your Digital Audio Recorder” and “Additional Dragon Preparations (Optional).”
If you have ODMS R6.4 Patch 0016 installed, you can also pair with these versions of Dragon:
- Dragon Professional Individual 14
- Dragon Professional Group 14
- Dragon Legal Individual 14
- Dragon Legal Group 14
- Dragon Professional Individual 15
- Dragon Legal Individual 15
Of course, for background voice recognition, you’re going to need something to record onto. Allow us to recommend the Olympus DS-7000 digital voice recorder.
Finally, speech recognition integration is only available on the Windows® platform. There is no support for this on the Mac® side.
How Does It Work?
In most iterations of the integrated speech recognition/dictation set up, the author creates dictations on a DS-7000. Then they dock the device, the Olympus software opens, and their dictations are downloaded. This workflow is also supported using the Olympus Dictation App, although files will be delivered via email, rather than downloaded.
When properly configured, the new dictation files hop into a queue, where they are then processed by Dragon. When the speech recognition step is finished, each of the audio files has a document attached to it containing the auto-transcribed results.
From there, either the author or transcriptionist opens the correction window, makes any necessary edits, and then exports the document in its final form. Any corrections made to misrecognitions by Dragon are integrated into the profile when adaptation is run.
This is known as background voice recognition.
Traditionally speaking, there are two approaches to adding Dragon speech recognition to your dictation-transcription workflow: either author-side integration or transcriptionist-side. By integration, we mean which side of the workflow you’re going to install Dragon on—and subsequently connect to the Olympus software.
Depending upon the number of authors and transcriptionists, there are strategies that work better in some scenarios than others. It usually comes down to three controlling factors:
- Real-time voice recognition
- Workload distribution
- Dragon profile access
Notice that we didn’t mention licensing as one of the factors, as the above editions of Dragon require a license for each author. Installing in one place or another won’t change that requirement.
So then, let’s explore the effective benefits and drawbacks of several possible schemas.
1 Author, No Transcriptionist
In this first scenario, the author is on their own. They have the ODMS Dictation Module and Dragon NaturallySpeaking both installed on their computer.
There is a lot of flexibility associated with this setup. Authors can create dictations on their handheld recorder, dock the device, and then allow it to run through speech recognition. They can then make any corrections needed and produce a final document.
Having Dragon installed means that they can also perform real-time speech recognition, with local access to their profile.
1 Author to 1 Transcriptionist
If you install Dragon and the ODMS Dictation Module on the author’s computer, he has the option of performing both real-time and background speech recognition. In addition, the profile can be stored locally.
In this configuration, you’d need to install the ODMS Transcription Module on the transcriptionist’s machine for the purposes of receiving files automatically and managing corrections.
If you only install Dragon and the ODMS Transcription Module on the typist’s machine, all files can be run through background speech recognition and have corrections integrated locally. Everything’s happening in one place.
The author can still have the ODMS Dictation Module for downloading and sending files, but he won’t have the option to perform either real-time or background speech recognition on his machine.
2+ Authors to 1 Transcriptionist
Having multiple authors changes the workload factor significantly. With each additional author generating audio files, it’s a definite advantage to have them running background voice recognition on their own dictations. Otherwise, the task falls to the transcriptionist, whose computer may not be able to process that many files in a timely manner.
This means that each author needs Dragon installed—and should be using the ODMS Dictation Module. Again, their profiles can remain local. As before, the transcriptionist needs the ODMS Transcriptionist Module.
Installing Dragon on one transcriptionist’s computer will keep your authors from having to install Dragon at all. When you have more than one installation to maintain, that becomes a definite advantage.
All profiles can be stored locally on the transcriptionist’s machine, which allows her to run dictations through background speech recognition using the Dragon-enabled Transcription Module. Corrections can be processed locally as well.
Naturally, authors can use the Dictation Module to download and deliver their dictations to the transcriptionist. For a lighter footprint, you can also simply bring the DS-7000 to the transcriptionist and have the files download directly into the speech recognition queue.
2+ Authors to 2+ Transcriptionists
In this particular schema, some of the apparent benefits of choosing integration sides start to flatten out. It’s a good idea to get some advice before implementing a speech solution in this environment. Be sure to contact your reseller, your IT department, or one of our expert technicians.
Let’s revisit our controlling factors:
- Real-time voice recognition:
If the authors need this functionality, then Dragon must be installed author-side.
- Workload distribution:
In this case, choosing between the authors’ and the transcriptionists’ machines depends heavily upon the amount of work to be processed and the computing power available.
- Dragon profile access:
Roaming is the ideal setup here, unless certain transcriptionists correct certain authors’ work, and there is no request for real-time speech recognition.
ODMS Settings and Dragon
Here are a few things to remember when using ODMS and Dragon together –
- Be sure that the DS-7000 recorder associated with a particular author has the correct author name assigned to the device. See “Setting Up the Dictation Module Software” for instructions.
- Associate the right author name with the right Dragon profile name in the voice recognition settings of ODMS. This is slightly different between the Dictation and Transcription Modules.
- If necessary, change the author name of existing dictations in ODMS by right-clicking and choosing “Edit Job Data.”