Start a simulated call
For demos and pipeline testing: two realtime voice agents hold the whole conversation — an interviewer and a scripted persona, and the pipeline processes it exactly like a real recording. Each persona is written with known expected outcomes, so the draft can be checked against ground truth.
The conversation streams live over the session’s SSE stream (turn.start / turn.delta / turn.audio / turn.done), paced to natural speed.
Authorizations
Sign in from your terminal with the OAuth device flow (see the Authentication page for a copy-paste block), or grab your current session token from https://console.silmo.ai/api/auth/token while signed in to the console. Access tokens are short-lived (a few minutes) — refresh with your refresh_token. Everything a session produces is stored and served strictly within your organization's namespace.
Body
A persona id from GET /meta.
checklist = screening · dst = full assessment · fast-track = urgent pathway
checklist, dst, fast-track Read the persona's referral evidence pack into a briefing first, so the interviewer goes into the call already briefed and references the records by name during the call.
Chaining: the id of a completed, positive Checklist session. Its results are injected as referral evidence for this DST.
Hard cap on conversation turns. Defaults to a realistic per-tool budget.
Generate a short showcase call (~5 minutes): tighter turn budget and pacing overrides for both agents.
Mark the finished session as a pre-built template, surfaced on the console's start page.
Engine for the post-call diarized transcription pass.