Fdesser

I was having the same problem using Dial-dictate 4.0.6 .... We recently upgraded to the 4.0.9 and the random disconnects seem to have significantly diminshed, if not completely disappeared. A few other things to try for the D/4PCI ( I have the same card): In Dial-Dictate: Disable Software Based end of call. Set the end of call detect delay as low as possible... this may add dead air but wont cut the call (mine is set to 500). Reduce the DTMF tone detect delay - may add tones to recordings if its too short (mine is at 150). In DCOM configuration manager (you must stop dial-dictate before configuring the modem): Select the D/4PCI under: Configured devices=>D/x1D-PCI=>D/4PCI in slot xx. Double-click to open the properties. Select the Misc. tab and set "Disconnect tone" to No This seems to have resolved the issues for us. Not sure if it will, or will not, do the same for you but it may be worth a shot.

You can also manually delete the .wav files in the "c:\documents and settings\all users\nch swift sound\scribe\done" folder. I have found that scribe tends to "forget" to delete files sometimes. It is not reliable when it comes to removing the "done" data files. Also, I am not sure, but I believe that you need to leave the .txt and .dat files as pointers for completed work. The .wav files can be completed freely and, as these are the only files taking more than 1k of space, are really the only ones that must be removed.

refer to my posting :http://nch.invisionzone.com/index.php?showtopic=5445 You can adapt the location in the registry file to reflect a local drive as well.. you will have to duplicate the directory structure on the drive for it to function. I am not sure how it will effect incoming files, but you could manually move the exploded files from (if you are moving to D:) C: to D: I am not sure if that makes sense but I think it does If it doesn't respond and I will try to clarify. YES. ES, for some reason, will always make a .wav file in its working directory. Only way I could see to avoid this is to place the .wav file in the working folder for scribe and, somehow, manage to change the filename to the next number express scribe is going to assign, although this does not seem to be numerical order. Here Here! Or better yet, allow me to choose where I want in installed AND where I want the data to be stored.

Not sure, but have you tried SWITCH ? http://www.nch.com.au/switch/index.html It isnt listed as a supported type but it may be that PC Dart is a compressed .wav or something to that effect. I am not sure but it may be worth a shot.

So here is the scenario: We have 8 people typing incoming dictation from our accounts and all of them are within the office. In order to maintain a current file list, which is shared by all of the users, it was necessary for all users to see the same "file location" The initial solution was to run express scribe on a terminal server and have all users access the "same" copy of scribe from the TS virtual desktop. The advantage of this was that all of the users were pointing to the same source directory of "c:\documents and settings\all users\application data\nch swift sound\scribe\..." The disadvantage is that Terminal Services, for unknown reasons, caused a latency in the release of the foot pedal which accounted for a 4.5-5 second delay of streaming sound. In other words, you release the pedal and sound continues for 4.5-5 seconds. Not an easy thing to deal with, as a typist. So... a new solution needed to be reached. Here it is, quick, relatively simple, and it doesnt seem to cause much of a problem in a shared environment. The downside may be seen if Word or another windows based software package is used and the user has customized settings for their environment that are saved in the common file location. The end result of the "fix" is a registry change that will affect all programs which refer to the "all users\application data" registry entry. We did not have issues with this as we use WordPerfect 5.1 for DOS, imho the fastest proram, still, for power typing. Anyways, here are the steps which I took: Install Express Scribe onto the distribution server. Map network drive to the shared directory location. Create a share on your server to the "C:\Documents and settings\" folder. Map a network drive to this share. In this example we will use N:\ for the network drive. Create a batch file to modify the registry changes. Use notepad to create a new batch file (we will call it scribe.reg). Place the following entries into this file: [HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Explorer\Shell Folders] "Common AppData"="N:\\All Users\\Application Data" Place this file (scribe.reg) in N:\ Create an express scribe startup batch file. Use notepad to create a new batch file (we will call it scribe.reg. Place the following entries into this file: %systemroot%\regedit.exe /s N:\scribe.reg c: cd "\Program Files\NCH Swift Sound\Scribe" Scribe.exe Copy this file to C:\ AND to N:\ Create an express scribe shortcut to the batch file. Refer to the batch file on C:\ Change the icon (if desired) to use the icon in file Scribe.exe Thats it! Ultimately, it is transparent to Scribe. One more warning: make sure that scribe is set to use "Default sound out" or the users will not have control of their system sound. Also, changing the speed/sound in the program effects all users as the settings.ini file is shared by everyone. The net result of the above is to launch a small script that will incorporate the registry entry each time you run scribe, in order to ensure the list is always accurate. You must incorporate the .reg file into the registry AFTER the drive is mapped as Windows will rewrite the registry entry to C:\ if the original mapping is not present (and it isn't because windows maps the drive after loading the registry) when the registry is loaded. You may not be able to install scribe to an alternate location but you can spoof it into using a different location for data. I don't know if anyone else has experienced this, but hope this helps.