How it works
Voice-to-NERIS: file a fire incident report by talking
Voice-to-NERIS reporting means the incident report starts as speech instead of typing: a firefighter narrates the call, software turns the narration into a structured NERIS incident report, and the report is reviewed by a human before it is filed to the federal API. The narration is the input. The review is the safeguard. The API filing is what makes it a real report rather than a note.
You talk
Record on a phone, tablet, or laptop — in the bay, in the truck, at home. Tell the call the way you would tell another firefighter: what you were dispatched to, what you found, what you did, how it ended. It is built for how people actually talk, including the corrections.
It fills in the report
The recording is transcribed and turned into a structured NERIS report: incident type (all three levels of the hierarchy), timestamps, units, actions, and both narratives. Where we have a department’s local street and place names, the transcriber learns them before it listens.
You review
Every field the system filled in is visibly marked, so you always know what came from you and what came from it. If a required field is missing, it asks one question at a time. You correct anything it got wrong — this step is not optional and is not meant to be.
You file
One tap sends the report to NERIS through the official federal API — not a CSV export someone uploads later. If NERIS rejects something, you get the reason in plain language, pointed at the exact field. There is no auto-submit; nothing leaves without your explicit action.
The honest limits
- It can mishear and it can be wrong. Transcription and extraction are imperfect; that is exactly why every AI-filled field is marked and the review step exists. You are signing the report, so you read it first — same as it ever was.
- You still have to tell the story. If you do not say what units responded, no software should invent it. Missing required fields come back to you as questions, not as guesses.
- Filing requires your department's say-so. Like any software that files through the NERIS API, your department's NERIS Admin has to enroll our Client ID (Account → Enrollments, neris.fsri.org) before reports go out under its name. The no-account trial lets you test the voice-to-report part first; trial reports are kept separate and cannot be filed to NERIS.
That is the whole idea. If it sounds like it would save your evenings, here is the page about the evenings; if you are comparing options, here is how to evaluate NERIS software, including FireMic.