Create voice commands in plain words

Describe a command in plain words and use it straight away

Voice Command BuilderBy Corti
Meet with an AI expert

Book a call to scope it against your data and tools.

Overview

Let clinicians create voice commands and auto-texts in plain words. They say what they want, an agent builds the command, and they can test it by dictating straight away. No configuration forms.

How it works

The clinician describes what they want, for example “when I say normal knee exam, insert this text”. An agent turns the description into a command with trigger phrases and an action (insert text, go to the next section, delete a sentence), saves it, and the clinician can test it by dictating straight away. Inserted text is exactly what the clinician saved. The agent does not add findings of its own.

What Corti handles

  • Recognizing trigger phrases inside continuous dictation
  • Commands with variables, such as “go to {field}”
  • Medical speech recognition
  • Hosting in the EU or US

What you build

  • The command library screen
  • The actions your editor performs
  • Storing commands per user or per clinic

Make it yours

  • Choose the language people speak
  • Let punctuation be added automatically, or only when someone says it
  • Set how dates, times and numbers are written
  • Teach it names, places and local terms it should recognize
  • Swap abbreviations for your house style in the final text
  • …and much more

Frequently asked questions

How does a clinician create a command?

They describe it in plain words, such as “when I say normal knee exam, insert this text.” An agent turns that into trigger phrases and an action and saves it. The clinician can then dictate the trigger straight away to test it.

What can a command do?

Insert saved text, go to the next section or delete a sentence. Commands can also carry variables, such as “go to {field}”.

Could it add findings the clinician did not write?

No. Inserted text is exactly what the clinician saved, and the agent adds no findings of its own.

Where does this sit next to dictation?

Commands run inside dictation, so Medical Dictation comes first if you have not built it. Dictation Vocabulary Learner covers the other half: commands handle the phrases, and it handles misheard words and names.

What do we still have to build?

The command library screen, the actions your editor performs, and storage of commands per user or clinic. Corti recognizes trigger phrases inside continuous dictation and provides the medical speech recognition, with hosting in the EU or US.

How do we get started?

Use the Meet with an AI expert button to talk with a Corti expert about your editor and the actions it needs to support.

Add a voice command builder to your product

Book a call to see Voice Command Builder running on your own data and workflows.