Published
Agents Update: Speak What You Want the AI to Do
1 min read
HealthTasks Agents now support dictation. Speak your request, refine the wording, then submit. Faster prompting, lower friction, and a more accessible way to work with AI.
Typing a long Agents prompt is useful. It should not be the only way in.
We've added dictation to HealthTasks Agents so faculty and administrators can speak what they want the AI to do, then refine the wording before they submit.
What changed
You can now dictate into the Agents prompt instead of typing every word from scratch.
Speech becomes editable text in the same input. Clean up phrasing, add a detail, or tighten a sentence. Submit when the prompt looks right.
In practice:
- Speak the request. Describe the analysis, report, or workflow you want the agent to run.
- Edit before you send. The transcript stays visible so you keep control of the final wording.
- Submit when ready. Same Agents workflow. Less keyboard time up front.
Why this matters
Dictation is a UX improvement with practical upside:
- Faster. Speaking a request is usually quicker than typing the same instruction out.
- Lower friction. When the intent is already clear in your head, voice gets it into the prompt without slowing you down.
- More accessible. Voice input helps users for whom sustained typing is harder, slower, or less comfortable.
- Still editable. Dictation does not lock you into the first pass. Refine the words, then send.
Modern AI tools should meet people where they already work. Voice is part of that. Typing remains available whenever precision is easier at the keyboard.
Get started
If your program already uses HealthTasks Agents, dictation is ready in the prompt input. If you are evaluating how agentic clinical education software should fit faculty and admin workflows, we would like to show you the product.