Although I am using DPG 15.61, the results should be the same when using DPI 15.61. If I set my Auto-Formatting to ‘Numbers, if greater than or equal to 0’ then this is what I get when dictating the following.

”I want it one time only”
”I want it to times only”
”I want it free times only”
”I want it for times only”
”I want it 5 times only”

You can see that numerals only kick in on number 5. The first 4 examples are taking literal words instead.

Yet if I say 1.86, or 1% for example, it gives you the figures as it is supposed to.

I believe that Nuance has overlooked this fault. If everyone else who has upgraded to DPG or DPI 15.61 try this experiment, perhaps they can come back and tell us the results.

I am also getting the regurgitated problem when changing the Auto-Formatting to Numbers, if greater than or equal to 10.  The Auto-Formatting changes this to Numbers, if greater than or equal to 2 by itself. I don’t know if this is after a time, or after the next time you log into Dragon. I have to keep changing it to Numbers, if greater than or equal to 10 but it does not stick.

Somebody said that this had been fixed but it hasn’t with my copy so this is another instance that Nuance has missed.

Cheers

-------------------------

Tiger Feet

| DPG 15.61 | KnowBrainer 2017 | Windows 10 Professional /64 Bit | Intel® Core™ i9 Ten-Core Processor i9-10900K (3.7GHz) 20MB Cache |  32GB RAM. | 250GB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 2300MB/W) Boot Drive | 1TB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 3300MB/W) Storage Drive | Sennheiser D10 USB Wireless Microphone

If your Auto-Formatting Options look the same as ours, you shouldn't have this issue. Your user profile either has a wrong setting or is slightly corrupted. There is probably a real solution but the sledgehammer approach is likelier simpler. Try creating a new profile and then import your current user vocabulary and commands. Note that in this situation, you should be able to use the DragonBar Manage Vocabularies\Export feature which will preserve your training, specialty vocabulary and even your deleted words.

 

-------------------------

Forum Mission Statement
Trial Downloads
Dragon/Sales@KnowBrainer.com 
(615) 884-4558 ext 1

Sorry but

"I want it one time only"
"I want it to times only"
"I want it free times only"
"I want it for times only"

are misrecognitions, not auto-formatting hiccups. Auto-formatting handles whether a number is spelled out as a word or as a numeral. If Dragon doesn't properly recognise the number as such, no auto-formatting rule will ever kick in.

A new user profile is probably the best option but as I just wrote in another post, don't export your vocabulary, but only your custom words, preferably in XML format. The vocabulary is your hotel room, the custom words your suitcase. You don't take the whole room with you if you move, especially if you move because the window is broken.

Hope that helps, Stephan

-------------------------

www.egs-vertrieb.de - www.spracherkennungscloud.de

Nice analogy

+1 from me

 

mav

At least two of these are funny misrecognitions, however, due to homophones. Dragon ought to know better than to think you probably want to say "I want it to times only" or "I want it for times only".
Acoustically, these misrecognitions are dead accurate, but Dragon is doing a poor job of applying the language model. This is one of my pet peeves with version 15. Although its accuracy is very high, it makes a fair number of mistakes like this where it puts together words that really shouldn't show up in combination in any decent language model.
As some of us have discussed here before, it seems that version 15 overweights the acoustic model and underweights the language model. I wonder if moving the speed versus accuracy slider to the right results in greater weighting of the language model.
Put another way, be sure to ask at the front desk if your room is next to an elevator.

Yes, I know I can say, numeral before a number but I shouldn't have to.

Okay, so I tried:

I counted one item
I counted 2 items
I counted 3 items
I counted 4 items

Using your context, 3 of them corrected themselves but the 1st 1 did not. I tried it 3 times but I am still getting;

I counted one item, I counted one item. I counted one item I counted one item. Using your context.

So I suppose the next thing I should try as suggested by Stephan is to export my custom words and phrases in XML format, then reimport them? Do I need to go through the words and phrases first and take out any erroneous spellings?

Cheers

-------------------------

Tiger Feet

| DPG 15.61 | KnowBrainer 2017 | Windows 10 Professional /64 Bit | Intel® Core™ i9 Ten-Core Processor i9-10900K (3.7GHz) 20MB Cache |  32GB RAM. | 250GB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 2300MB/W) Boot Drive | 1TB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 3300MB/W) Storage Drive | Sennheiser D10 USB Wireless Microphone

I have always wondered why Nuance would choose to allow the user to allow number formatting "if greater than or equal to 2". The "2" allows for 2 many missed recognitions (as has been just documented with my dictation here). Why not just simply start with 3 or greater?

-------------------------

Bill
DPI 15.61, KB 2017, SpeechStart +, customized Desktop PC (AMD 12 core 9th gen Ryzen 9 3900x@3.8 GHz, 32 GB DDD4 SSD), Crucial 1 TB m.2 PCIe NVMe SSD HDD, Windows 10 Pro 64-bit, Philips SpeechMike Air 4000, Philips speech Mike 3500, SpeechWare 3-in-1 TableMike, MS Office 365 Professional and Microsoft Surface Pro 2017, Windows 10, 512 GB SSD, 16 GB ram, i7

I shall have to do some fiddling about and see what happens. As I say it is not the end of the world. Just a niggling PIA.

Is the custom words and phrases scenario the same as the other problem I mentioned? That being having the formatting set on 'numbers, if greater than or equal to 10' and then the next time you log into Dragon, it has changed by itself to 'numbers, if greater than or equal to 2?' Am I the only person this happens to?

Cheers

-------------------------

Tiger Feet

| DPG 15.61 | KnowBrainer 2017 | Windows 10 Professional /64 Bit | Intel® Core™ i9 Ten-Core Processor i9-10900K (3.7GHz) 20MB Cache |  32GB RAM. | 250GB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 2300MB/W) Boot Drive | 1TB SAMSUNG 970 EVO PLUS M.2, PCIe NVMe SSD (up to 3500MB/R, 3300MB/W) Storage Drive | Sennheiser D10 USB Wireless Microphone