> ## Documentation Index
> Fetch the complete documentation index at: https://api-docs.recnote.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Full-file flow

> Send one complete recording, get the note back.

Use this when you have the entire consultation recording available.

## Generate a note from audio

```bash theme={null}
curl -X POST https://api.recnote.ai/api/v1/notes \
  -H "Authorization: Bearer $RECNOTE_API_KEY" \
  -F "file=@consultation.wav" \
  -F "language=sv"
```

The server transcribes the audio and generates the note in a single call,
returning both:

```json theme={null}
{
  "transcript": "...",
  "note": { "cause": "...", "subjective": "...", "...": "..." }
}
```

## Supported audio

* Formats: `wav`, `mp3`, `m4a`, `ogg`, `webm`.
* Max body size: 500 MB. Large files are split and transcribed in parallel
  server-side.

## Transcript only

If you only want the text and will generate the note later (or not at all):

```bash theme={null}
curl -X POST https://api.recnote.ai/api/v1/transcriptions \
  -H "Authorization: Bearer $RECNOTE_API_KEY" \
  -F "file=@consultation.wav" \
  -F "language=sv"
```

```json theme={null}
{ "transcript": "..." }
```

## Note from an existing transcript

Already have the text? Skip audio entirely and POST JSON:

```bash theme={null}
curl -X POST https://api.recnote.ai/api/v1/notes \
  -H "Authorization: Bearer $RECNOTE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "transcript": "Patienten söker för huvudvärk sedan tre dagar...",
    "language": "sv"
  }'
```