Voice Recognition Fact Sheet


Information for Doctors

  • Typist Mode (Typist.png) will allow doctors to dictate a letter into text and save the recording directly within the Patient’s file for the admin staff to edit. The referring and interested parties will be automatically populated into the default letter template. In Typist Mode, the VR updates the dictionary once the admin has edited the text and selected ‘Create letter from recognised text’.
  • Self Mode (Self.png) will show the transcription of the VR dictation in real time on the screen as the doctor dictates. The doctor then corrects their own dictation, however, this is usually less practical as the editing must be done in the single sitting to maintain Voice Recognition accuracy. 

  • In Self Mode, the VR updates the dictionary once the doctor has completed the dictation, edited the text, and selects ‘OK’.

We recommend the use of Typist Mode and having admin staff check and create letters on the doctor’s behalf to maintain Voice Recognition accuracy.


  • Building Voice Recognition accuracy takes time and consistency. After editing and creating approximately 2 hours worth of dictation, the VR accuracy will be optimal. 


  • Repetition, Repetition, Repetition. VR will learn your preferred syntax style, which it frequently hears through repeated sentence structure within your dictations. The VR will add new words to your custom dictionary after the 3rd time it is used. 


  • Clinic to Cloud Voice Recognition uses medical terminology. Focus your dictations on diagnosis/symptoms/assessment. Avoid dictating patient names, as the VR is not designed to learn patient names and can lead to dictation inaccuracies. 


  • When dictating into the Clinic To Cloud App, please note that the auto-correct function may change spelling or syntax, and if saved without checking, VR may learn incorrectly. Please be sure to double check your dictations if self-correcting. Clinic to Cloud voice recognition uses UK English medical dictionary, however please take note; most auto-correct systems use US English.


  • View the Voice Recognition dictation commands here.


Information for Admin Staff

  • Correct mistakes individually. When correcting the VR, it’s important not to erase all information and rewrite the entire paragraph, as this confuses the learning algorithm and can decrease the accuracy of the VR. Instead, erase and correct individual words or grammatical errors individually. 


Frequently Asked Questions


Question Answer
How much does Voice Recognition cost? A Voice Recognition user licence costs $150 per month per user, added on to an existing Clinic to Cloud doctor user subscription. To add a Voice Recognition licence to your account, fill in the form here.
Is everyone charged separately for this service? Yes, a Voice Recognition user is an individual subscription per doctor user.
We use a lot of repeat information in subsequent letters. For example, past medical history and current medications. Is there a way or retaining that information with appropriate boxes? This would be best handled by adding a Table to your Letter Template with the required number of columns and rows. You can then have a header row with the name of the fields, and then in the row below, place the Tag you'd like to use (eg Past Medical History, Family History, or Clinical Information). The Current Medications field can be imported in a table format automatically, and so will not need to be a part of the table mentioned above. See this article for more information.
What happens with dictation if you stumble and say um ahh etc Once your Voice Recognition service has been trained, it will pick up on nuances in your speech such as your pauses and "umms" and "ahhs" and know to exclude these, as they will be removed from your text when the recognition is corrected manually by yourself or your team over time. You may also choose to say "scratch that word" or similar during dictation if you find yourself using these terms at the time. 
How does Voice Recognition manage accents? Your Voice Recognition service will begin to recognise your accent and manner of speaking and will improve the quality of your recognised text over time through training the system. 
If I am using Siri / voice recognition on my iPhone with Clinic to Cloud Doctor App, does that content then get exported to iCloud? All information entered into Clinic to Cloud is hosted within Australia and will not be exported outside of Australian borders. However, any recordings made using Siri or other inbuilt smartphone Voice Recognition software are often stored on overseas servers, so we strongly advise not dictating any clinical information into voice recognition to be compliant with Australian data privacy legislation.
Can you use a combination of dictatation and shortcuts in the clinical consult window? You can use a combination of both dictation and shortcuts as you would like when writing and dictating consultation notes. You may click and begin recording your dictation, and once this has imported into the field you selected, you could then begin typing the beginning of your Dictionary macro to insert your shortcut. 
When is the new letter editor due to be released? Our development team is currently in the final stages of releasing a new letter editor with improved formatting and usability features. We will notify all customers when the new letter editor is available.
Do you need a special microphone to dictate into Clinic to Cloud accurately? No, you do not require a specific device for dictating into Clinic to Cloud. You may choose to use the inbuilt microphone function in the computer (if there is one), an external microphone or headset (such as one that you would use with your mobile phone). Clinic to Cloud sells dictaphones for those who prefer to use one, please enquire at support@clinictocloud.com.au
Can you dictate directly into the consult boxes instead of using short cuts? It is possible to dictate directly into the consult fields in the letter by placing your cursor into any field where you can see the fire symbol (indicates dictation capability), clicking on the Microphone at the top of the screen, and then beginning your dictation.
How many hours of dictation do you think a doctor needs to dictate for Voice Recognition to be accurate? Before commencing training, we request that you provide us with a minimum of 20 clinical letters for us to build your Voice Recognition profile with. Thereafter, we recommend 2-4 hours of dictating to teach the voice recognition system your accent, nuances, tone and word choice etc.
What happens if I want to change something in my letter after I’ve dictated it? Will it negatively affect the voice recognition learning? This will depend on where you make the correction from. If you have opened the dictation and corrected the recognition and created the letter from that recognition, then changes made after then will not teach the recognition system. It will not negatively affect the system, but it is within the corrections window that you must make the changes in order for the recognition service to learn.





Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request