Made in 10Docs
GuidesAPI
ReferenceCalls

Get a call

One call with its transcript, its analysis and the tools it used.

GET
/calls/{call_id}

Authorization

secretKey
headerAuthorizationBearer <sk_live_… or sk_test_… (a prefix and 40 base62 characters)>

A secret key from the dashboard, sent as Authorization: Bearer sk_live_…. Live keys run real, charged calls; test keys (sk_test_…) run calls of at most 60 seconds, 20 a day per organization, never charged. Each key sees only its own env's calls, meetings, webhooks, memories and analytics. Publishable keys (pk_…) are refused.

Path Parameters

call_id*string

The call's id.

Match^call_[0-9A-Za-z]{20}$
Example"call_2Gx7Mb9KpQ4sVn1ZrT6e"

Response Body

The call.

application/json
  1. response

One call, with what came of it.

Example
{  "id": "call_2Gx7Mb9KpQ4sVn1ZrT6e",  "character_id": "chr_4fT9kQ2mXw7LbN3pZs8V",  "character_name": "Alex",  "status": "ended",  "source": "api",  "env": "live",  "max_seconds": 300,  "billed_seconds": 94,  "charged_usd": 0.94,  "price_per_min_usd": 0.6,  "end_reason": "completed",  "error": null,  "metadata": {    "ticket": 7  },  "created_unix": 1793300000,  "settled_unix": 1793300100,  "mode": "video"}
id*CallId

A call's id (ours; the engine's session id is never shown).

Match^call_[0-9A-Za-z]{20}$
Example"call_2Gx7Mb9KpQ4sVn1ZrT6e"
character_id*CharacterId

A character's id.

Match^chr_[0-9A-Za-z]{20}$
Example"chr_4fT9kQ2mXw7LbN3pZs8V"
character_name*|

The character's name as the call started.

status*CallStatus

starting (for a moment, while the engine starts it), waiting (started: the browser may connect, and stays so while the call runs), ended, failed (our side; nothing charged).

Value in"starting""waiting""ended""failed"
source*CallSource

Where the call was started (meeting, a character sent into a meeting).

Value in"dashboard""api""widget""meeting"
env*Env

live (real, charged) or test (a test key's, never charged).

Value in"live""test"
max_seconds*integer
billed_seconds*integer

The seconds the call counts for: from the connection to the end, rounded up, at least 30, at most max_seconds. 0 until it settles, and for a call that failed, never connected or never played a reply. Test calls count seconds but are charged nothing.

charged_usd*Usd

US dollars.

Range0 <= value
Example0.94
price_per_min_usd*|

The price it ran at; null for test calls and free organizations.

end_reason*|

Null until it settles.

error*|

A plain sentence for a call that didn't go as planned ("The call couldn't start. Nothing was charged.").

metadata*

What you sent with startCall ({} otherwise).

created_unix*Unix

Unix seconds, UTC.

Formatint64
Example1793300000
settled_unix*|

When it was settled (charged); null while it runs.

Formatint64
character_version?|

The character's published version the call ran on. Null for a dashboard talk on a draft, and for calls made before versions existed.

mode?|

What the call ran in: video with the character's face, voice without one. Null for calls made before 2026-11-12.

Value in"voice""video"null
transcript*array<>

The turns. On an interviewer character's call no turn has delivery.

chat*array<>

What was written in a meeting's chat, in order; empty for any other call.

talk_numbers*|

The engine's numbers about how each person talked; null when it saved none, and always null for a call whose character is an interviewer now or was when the call was made.

character_role*|

The character's role as it is saved now (interviewer, support, …); null once the character is deleted.

live?boolean

Only while the call runs and the engine answers it; whether the person's side is connected.

analysis*|

What came of it after the call (null only for a call whose character can't be found).

tool_calls*array<>

The tools it used (your own and your MCP servers'), oldest first.

recording*

A call's recording as the call shows it. pending (the call is on, or has just ended and its file is being finished), ready (it can be listened to: getCallRecording), none (nothing was said: the call never connected), off (the character's record was off), deleted (by you, or when the organization's days to keep recordings, 30 unless we changed them, were up).

Example
{  "status": "ready",  "seconds": 94,  "bytes": 6016044,  "deleted_unix": null}
diagnostics*

Why a call felt slow or odd, from the timings the engine keeps for every turn: each stage at p50, p95 and max (nearest rank), each turn's stages, and hints in plain English. Only numbers, never words: they stay when the transcript is deleted.

Example Request
curl -X GET 'https://api.daxov.com/v1/calls/call_2Gx7Mb9KpQ4sVn1ZrT6e'
Example Response
{  "id": "call_2Gx7Mb9KpQ4sVn1ZrT6e",  "character_id": "chr_4fT9kQ2mXw7LbN3pZs8V",  "character_name": "Alex",  "status": "ended",  "source": "api",  "env": "live",  "max_seconds": 300,  "billed_seconds": 94,  "charged_usd": 0.94,  "price_per_min_usd": 0.6,  "end_reason": "completed",  "error": null,  "metadata": {    "ticket": 7  },  "created_unix": 1793300000,  "settled_unix": 1793300100,  "mode": "video",  "transcript": [    {      "n": 1,      "customer": "",      "ai": "Hi, I'm Alex, an AI assistant. What can I help you with?"    }  ],  "chat": [    {      "at_ms": 0,      "from": "string",      "text": "string",      "to": "everyone",      "addressed": true    }  ],  "talk_numbers": {    "people": [      {        "who": "string",        "turns": 0,        "words": 0,        "speaking_ms": 0,        "share": 0,        "wpm_median": 0,        "wpm_min": 0,        "wpm_max": 0,        "pauses": 0,        "fillers_per_min": 0,        "interruptions_made": 0,        "interruptions_received": 0,        "answer_after_median_ms": 0,        "rough": true      }    ],    "character_ms": 0,    "character_share": 0  },  "character_role": "string",  "live": true,  "analysis": {    "status": "ready",    "summary": "A lead asked about the Pro plan and booked a demo.",    "fields": {      "budget": 200,      "plan": "Pro"    },    "verdict": "success",    "reason": "They booked a demo for Tuesday.",    "error": null,    "analyzed_unix": 1793300110  },  "tool_calls": [    {      "id": "tcl_4Zq9Hv2NcP7xMb1KrW6s",      "tool_id": "tool_6Mx1Rb8QwT3kNv5LcZ9p",      "tool_name": "order_status",      "ok": true,      "status_code": 200,      "duration_ms": 412,      "error": null,      "created_unix": 1793300050    }  ],  "recording": {    "status": "ready",    "seconds": 94,    "bytes": 6016044,    "deleted_unix": null  },  "diagnostics": {    "status": "ready",    "reason": "never_answered",    "face": true,    "replies": 0,    "quick_replies": 0,    "interruptions": {      "talked_over": 0,      "cut_off": 0    },    "talk": {      "person_ms": 0,      "ai_ms": 0,      "ai_share": 0    },    "stages": [      {        "stage": "reply",        "n": 0,        "p50_ms": 0,        "p95_ms": 0,        "max_ms": 0,        "mean_ms": 0,        "max_turn": 0,        "good_ms": 0,        "slow_ms": 0      }    ],    "hints": [      {        "code": "face_never_started",        "level": "warn",        "text": "string"      }    ],    "turns": [      {        "n": 0,        "kind": "reply",        "at_ms": 0,        "reply_ms": 0,        "hearing_ms": 0,        "mind_ms": 0,        "voice_ms": 0,        "face_ms": 0,        "greeting_ms": 0,        "quick_reply": true,        "quick_reply_ms": 0,        "interruption": "talked_over",        "stop_ms": 0,        "person_ms": 0,        "ai_ms": 0      }    ]  }}