19.2.0 nVoq Dictation, Voice, Administrator, & APIs Release
September 2024 - See Production Deployments for deployment dates.
Changes listed on this page may be available on the Test system (test.nvoq.com), but there could be some that are not there yet.
Each item lists the version and build number in which it was tested. To see the version and build number currently deployed on Test, visit Test System Deployment Dates.
For release notes for nVoq mobile apps, see:
- nVoq Mobile Voice for Android Version History - 19.2.0
- nVoq Mobile Voice for iOS Version History - 19.2.0
- nVoq Wireless Microphone for Android Version History - 19.2.0
- nVoq Wireless Microphone for iOS Version History - 19.2.0
nVoq Dictation
The updates in this section impact all dictation clients, including nVoq Voice and Mobile Voice, as well as ISV partner clients using nVoq dictation.
- Fix for “x” Not Returning in place of “times” in STABLETEXT Dictations - nVoq dictation was returning the word "times" instead of "x" when that word was dictated before a number using STABLETEXT. For example, "times 30 seconds" would return in the transcript instead of "x30 seconds." The word "times" has been added to the list of special words (along with "plus," "minus," and "dash") that wait for the beginning of the next segment before committing to a format. (D-13460, 7/25/2024, 19.2.0 b29)
- Limit on the Number of Vocabulary Terms That Can Be Processed - Due to issues with profiles that have an excessive number vocabulary terms, user profiles will now only process up to 15,000 vocabulary terms in dictation. Anything over that will not be recognized. This limit should be more than enough for our partners based on current usage. The 15,000 limit on vocabulary terms includes a combination of both organization and account vocabulary. Terms are counted in the order that they were added. (D-13504, 7/29/2024, 19.2.0 b32)
- Mid-Dictation Command Accuracy Improvement - In previous releases a longer, multi-word mid-dictation command may not be recognized when there are pauses between words in the command. Work has been done to improve the accuracy of long mid-dictation commands when there is a silence between words. (B-17312, 7/31/2024, 19.2.0 b34)
- Improvement of Boundary Request Response During Silence - During periods of silence or non-spoken signaling (like hitting the tab key) BOUNDARYREQUEST messages now receive faster response via end of segment messages. This work made an improvement to the work previously done to allow a non-verbal user action to be acted upon in sequence with dictated text (see API release note for B-17701 for more information). (B-17735, 8/15/2024, 19.2.0 b44)
- Fix for Mid-Dictation Commands with Acronyms - Work was done to allow mid-dictation commands with acronyms to work. In the previous release, commands such as "CAD" and "IV" were not working at all. (D-13411, 9/6/2024, 19.2.0 b60)
- Fix for Dictations with Multiple Commands Return Incorrectly - When multiple mid-dictation commands are used in one dictation, word markers are dropped. This was causing commands to return incorrectly, such as the second command being overwritten with the first, etc. This has been fixed. (D-13566, 9/11/2024, 19.2.0 b60)
nVoq Voice
- Customize Paste Delay Settings for an Organization - nVoq staff can now change the default settings of the Note Assist Delay and the Paste Delay for an entire organization using the API. Settings changed via API apply to new and existing users unless the existing user has already changed their default delay. In that case, the user's customized setting remains. This will allow nVoq to increase paste delay settings for all users in an organization at any level at once so that each clinician does not have to manually change them in order for dictation or Note Assist to work consistently in a virtual desktop environment, like Citrix. This can be done for organizations now, you do not need to wait for 19.2.0 to be released. (B-17638, 7/8/2024, 19.2.0 b6)
- Introducing Intelligent Formatting - Voice has a new feature called Intelligent Formatting which can add punctuation to text without you having to speak it. Just dictate as you would normally speak, without punctuation. After you stop your dictation, tap the F9 key on your keyboard to open the Intelligent Formatting window and apply punctuation and proper casing, even correct a pronoun if you say one that does not match. It works on dictated or typed text, and is available to everyone. (B-17676, 8/27/2024, 19.2.0 b50)
nVoq Administrator
- Removed HTML Automation Shortcut from New Shortcut Menu - As part of our cleanup work to remove the information that was only needed for our retired SayIt dictation client, the HTML Automation shortcut option has been removed from the new shortcut menu on the Shortcuts page. Existing HTML Shortcuts have not been deleted, but the "Shortcut Designer" used to create and edit them has been removed. The code for HTML Shortcuts will open in a plain text editor on the Edit Shortcuts page. (B-17609, 6/20/2024, 19.2.0 b4)
- Removed SayIt Options from R&C Type Menu - As part of our cleanup work to remove the information that was only needed for our retired SayIt dictation client, the Microphone Setup Result and Sub-Shortcut options have been removed from the Type menu on the Review & Correct page. If these activities exist in the selected date range they will still appear in the list view when "All" is selected from this menu. (B-17609, 6/20/2024, 19.2.0 b4)
- Removed SayIt Settings from Add/Edit Accounts, Update Accounts, & Auto Provision Account Settings - As part of our cleanup work to remove the information that was only needed for our retired SayIt dictation client, "SayIt" has been removed from the Products list on the Add Account, Edit Account, Upload Accounts pages, and from the Products list on the Identity Providers page Auto Provisioning Settings tab. SayIt account options, including Voice Shortcuts, Fast Shortcuts, Show Paste History, Direct Text, and Automatic Spacing have also been removed from the Options list on the Add Account, Edit Account, Upload Accounts pages. HTML Shortcuts options were also removed from these pages as well. (B-17605,7/5/2024, 19.2.0 b6)
- New Keep Note Assist Transaction Text Setting for Groups - We are working on adding the ability to review text that was run against Note Assist audits for agency quality control and coaching purposes. As part of that work, a new setting has been added to the Organizations page for groups labeled Keep Note Assist Transaction Text. This setting controls how long the narrative text that a Note Assist audit is run against is stored. The default value is 365 days for new groups. For groups existing prior to the 19.2.0 deployment the value defaults to match what is set for Keep Text, but can be changed after deployment. (The two values do not have to be the same.) You can view the setting for multiple groups at a glance using Customize Columns to add the Keep Note Assist Transaction Text column to the Organizations page list view, and you can change the setting for multiple groups at once using the mass updates feature. Only text run against Note Assist audits in Voice and Mobile Voice is stored; narrative text run against batch audits and text used in testing audits within nVoq Administrator is not stored. (B-16572 & B-17710, 8/9/2024, 19.2.0 b40)
- View nVoq Voice Conversation Transcript and Feedback - When a dictation is recorded in nVoq Voice Conversation mode* there will be a third download link below the Corrected text box on the Detail Review & Correct screen. The Download Conversations Text link will download a JavaScript Object Notation (JSON) file that will include the diarized transcript of the conversation as well as any feedback that may have been submitted for the transcript from the Conversations window in Voice. JSON is a plain text file that can be opened in any text editor. *19.2.0 is a beta release for nVoq Voice Conversations. It is not feature complete, and must be enabled for your account or organization by nVoq via API. (B-17744, 9/13/2024, 19.2.0 b64)
Bug Fixes in nVoq Administrator
- Fix for Auto Provisioning Timestamp Updated with nVoq Administrator Changes - The Last Auto Provisioning Update timestamp (at the bottom of the Edit Account page for auto provisioned accounts) should only be updated when a change is made to the account via the SSO provider using our SCIM server. In previous releases this timestamp was being updated when changes were made to the account in nVoq Administrator too. This has been fixed. (D-13536, 7/5/2024, 19.2.0 b10)
- Fix for Page Unresponsive After Changing Password in nVoq Administrator if Account is 2FA-Enabled - When two-factor authentication (2FA) is enabled for an account, nVoq Administrator becomes unresponsive after the user changes their password either by clicking the person icon and changing it when already signed in, or by being automatically directed to the Change Password page while signing in due to an expiring (or expired) password. The work-around is to sign in again and the new password will work. In 19.2.0 a 2FA-enabled user signing in to nVoq Administrator for the first time or when the password is expired/expiring is redirected to the Change Password screen and then redirected to DUO 2FA after the password is successfully changed. A 2FA-enabled user opting to change their password when they are already signed in to nVoq Administrator will not have to re-authenticate through DUO once the password is successfully changed. (D-12928, 7/5/2024, 19.2.0 b11)
- Fix for Failed Results on Account Tracking Reports - In the previous release the Account Tracking Detail and Account Tracking Summary reports could send an incorrect message of "No report results" due to a failed query. This has been fixed in the 19.2.0 release. (D-13556, 8/6/2024, 19.2.0 b35)
- Fix for Existing Substitutions with Different Casing Fail on Upload - In the previous release you could not upload a substitution file if any substitution in the file already existed but had different casing (e.g. "nVoq" vs. "NVOQ"). This would happen even if the "Overwrite existing" toggle was enabled. When attempting to do this, the upload would fail with a message that the substitution already existed. This has been fixed. (D-13560, 8/15/2024, 19.2.0 b43)
nVoq APIs
There may be changes listed here that are not yet available on the Test system.
Each item lists the version and build number in which it was tested. To see the version and build number currently deployed on Test, visit Test System Deployment Dates.
- Removed SayIt From List of Available Products -As part of our cleanup work to remove the information that was only needed for our retired SayIt dictation client in nVoq Administrator, SayIt is no longer listed as an available product on the endpoint to retrieve a list of products. (B-17605,7/9/2024, 19.2.0 b6)
- Boundary Requests Allow Non-Verbal Events to Queue Actions in STABLETEXT Dictations - Multiple updates were made to WebSocket dictation methods in order to allow a non-verbal user action, such as pressing a foot pedal or a function key, to be acted upon in sequence with dictated text. First, a new subscription (
SEGMENTEND
) which instructs the server to send back end-of-segment messages has been added to the STARTDICTATION method. Second, a new optional parameter (requestedBoundaryId
) has been added to the BOUNDARYREQUEST method. This is an ID the requester may send to recognize an end-of-segment response associated with a BOUNDARYREQUEST. Finally, newENDOFSEGMENT
GET endpoints have been added to WebSocket Dictation and WebSocket Transcription Observer. This message indicates a point in a dictation where a pause in speaking occurs and a segment of speech is finished. Any boundary request applied during this segment will be returned within this end-of-segment message. No boundary request with arequestedBoundaryTime
later than thesegmentEnd
will show up. Instead, they appear in a later ENDOFSEGMENT message. All of these changes allow boundary requests to queue non-verbal events within dictation audio, so that the non-verbal user actions can be performed in sync with STABLETEXT dictations. (B-17701, 7/9/2024, 19.2.0 b18)- https://test.nvoq.com/apidoc/dictation_websocket#tag/WebSockets-Dictation/operation/wsMessageStartDictation
- https://test.nvoq.com/apidoc/dictation_websocket#tag/WebSockets-Dictation/operation/wsMessageBoundaryRequest
- https://test.nvoq.com/apidoc/dictation_websocket#tag/WebSockets-Dictation/operation/wsMessageEndOfSegment
- https://test.nvoq.com/apidoc/dictation_websocket#tag/WebSockets-Transcription-Observer/operation/wsMessageEndOfSegmentObserver
- Word Markers Added to HYPOTHESISTEXT - Word markers have been added to HYPOTHESISTEXT messages. To use them you must include a request for HYPOTHESISTEXTWITHMARKERS rather than HYPOTHESISTEXT. You cannot request both. If both are requested you will see a warning message, and HYPOTHESISTEXTWITHMARKERS will win. HYPOTHESISTEXTWITHMARKERS can be requested together with STABLETEXT. The STARTDICTATION WebSocket method has been updated to explain that HYPOTHESISTEXTWITHMARKERS behave just like HYPOTHESISTEXT except that they have markers populated for all completed segments. The TEXT WebSocket method has been updated to clarify that
audioStart
times are based on the audio stream, starting at zero. (B-17711, 8/1/2024, 19.2.0 b35)