I do have DMPE 4.31 that I have been using with very good if not excellent accuracy levels ( Phillips Speech Mike Premium Air 4000 on an HP windows Desktop Windows 11 ) 

I have recently transitioned to DMO cloud version and my accuracy levels have significantly deteriorated despite dictating the same way 

I noted there is a microphone setting which is set at default . I have tried it at default and at a few  different sensitivity settings but my accuracy is still not acceptabel 

I was wondering if anyone has had experience the same when transitioning and what would be the best setting for a speech mike microphone for DMO 

Kind regards 

In theory, your accuracy should be the same or possibly a little better. Unfortunately, we are hearing the opposite from a number of end users. Did you, perchance, disable DMPE 4.3.1's auto-gain control by removing the checkmark from Automatically adjust the microphone level as I speak (under the Check Microphone menu)? Disabling this feature in DMPE 4.3.1 has proven to be helpful for many physicians. Unfortunately, you do not have this option, along with many others, in DMO

Change "No" to "Know" w/KnowBrainer 2022
Trial Downloads
Dragon/Sales@KnowBrainer.com 
(615) 884-4558 ex 1

I used DMO for approximately one month, but finally gave up on it, because of the frequent need for corrections.

I returned to DMPE 4.3, which was extremely accurate using a trained profile.  In my experience, the difference between DMO and DMPE 4.3 was significant.

Out of curiosity: Did you stick with the same user profile all the time or have you tried resetting the user profile?

From our experience with Dragon Medical Server, it's very unlikely to deliver low accuracy when DMPE was working well, because deep down they're using the same technology.

What could have an influence, though, is the user profile.

With DMPE, your user profile (mainly the acoustic model) is adapted even during the very short period where you adjust the audio system.

While DMO doesn't have such a dialog, from what I heard the profile is also adjusted during the first few dictation sessions. If something went wrong there, you stick with the badly adapted profile and your accuracy will suffer all the time.

For this, Nuance has added an option to NMS to reset the speaker profile - similar to what you usually tell DMPE users to do once their profile goes wayward.

I'm not sure if a "regular" user can reset his profile, but at least your company's DMO admin or Nuance support should be able to do so.

hth

mav

This is a follow-up to my previous grief after transitioning from DMPE 4.3 desktop application to Dragon Medical One.

My accuracy with DMPE 4 .3 was almost hundred percent on the desktop application but dropped drastically after transitioning to the cloud version (DMO)

I tried playing around with two of my best microphones (the Philips speech Mike SMP 4000 and the Power Mic III Nuance microphone III) at the recommended sensitivity without success. I then tried multiple microphone sensitivities using the settings options, again without any success

After some frustration, I then went and purchased an AU$23 microphone (US$16) USB microphone. (USB headset) With this change alone, I am now managing 99+ or even hundred percent accuracy with my dictation.

I believe therefore that there must be a difference in the speech engines between the desktop application and the cloud version in the manner they handle microphones.

Regards everyone.

Now that you get good/excellent results with that cheap headset I think that confirms my initial guess: Your profile (namely your acoustic model) has been created with microphone characteristics of a low quality microphone.

If you're then dictating with a high qualilty microphone (both the SpeechMike and the PowerMike deliver very high quality), the audio characteristics do not match the ones stored in your profile and so the results are substandard.

I'm pretty sure you could have achieved the same results with a SpeechMike or PowerMike if you simply had reset your profile and then performed your first few minutes of dictation with a SpeechMike or PowerMike so that your acoustic model gets adapted for the high-quality mikes.

hth

mav

Thanks for the reply Mav.

In fact ,the reverse has happened. After doing a few excellent dictations with the 16 bucks microphone, i have simply now switched back to the Phillips Speech mike (which was originally giving low accuracy ) and now now Im getting my DMPE 4.3 equivalent accuracy levels with the Phillips Speech Mike.

Hopefully it remains so 

Regards

Ah, good to know.

Did you have your profile reset in the meantime?

mav

This is a further follow-up regarding my transition to Dragon Medical One from DMPE 4.3.

As per the sage advice of Mav , I explained my loss of accuracy to the administrator of the account.

He reported that some people have had reduced accuracy if their profile originally built up poorly.

As per Mav's observations and also the administrator's experience, resetting the profile and building it from scratch akin to creating a new profile altogether in DMPE-4.3 would fix the problem

I requested this total reset and started dictating normally as I do in the DMPE 4.3

I would like to say the accuracy is now amazing (almost hundred percent) except for misrecognition of a few similar sounding words rather than total miscues

It is a shame this reset cannot be done from the user's end 

Thank you Mav for saving me 

Regards 

Tapiwa