resources/live/subresources/sessions/methods/accept/index.md +0 −443 deleted
File Deleted View Diff
1## Accept call
2
3**post** `/live/sessions/{session_id}/accept`
4
5Accept an incoming SIP call. Supply session with type live, the model, and startup configuration. Before accepting calls, follow the [Live prompting guide](/api/docs/guides/live-prompting) to write frontend conversation instructions and a separate backend prompt. SIP media format is negotiated; omit audio.format.
6
7### Path Parameters
8
9- `session_id: string`
10
11### Body Parameters
12
13- `session: object { model, type, audio, 4 more }`
14
15 Model and startup configuration for the Live session that answers the incoming SIP call.
16
17 - `model: string or "gpt-live-1"`
18
19 The Live model to use for the accepted call.
20
21 - `string`
22
23 - `"gpt-live-1"`
24
25 The Live model. Required in the session configuration for every transport; do not pass it as a URL query parameter.
26
27 - `"gpt-live-1"`
28
29 - `type: "live"`
30
31 The session type. Always `live`.
32
33 - `"live"`
34
35 - `audio: optional object { output }`
36
37 Startup audio output configuration. SIP negotiates the media format; audio.format is only accepted for primary WebSockets. Voice cannot change after startup.
38
39 - `output: optional object { voice }`
40
41 Settings for speech generated by the Live model. Choose the voice before starting the session.
42
43 - `voice: optional string or "alloy" or "ash" or "ballad" or 19 more or CustomVoice`
44
45 The voice used for Live speech, as a built-in voice name or a custom voice object containing its ID. Defaults to `marin` and cannot change after startup.
46
47 - `string`
48
49 - `"alloy" or "ash" or "ballad" or 19 more`
50
51 The voice used for Live speech, as a built-in voice name or a custom voice object containing its ID. Defaults to `marin` and cannot change after startup.
52
53 - `"alloy"`
54
55 - `"ash"`
56
57 - `"ballad"`
58
59 - `"beacon"`
60
61 - `"bossa"`
62
63 - `"cedar"`
64
65 - `"cinder"`
66
67 - `"coral"`
68
69 - `"delta"`
70
71 - `"echo"`
72
73 - `"gleam"`
74
75 - `"marin"`
76
77 - `"meridian"`
78
79 - `"quartz"`
80
81 - `"ripple"`
82
83 - `"sage"`
84
85 - `"shimmer"`
86
87 - `"stone"`
88
89 - `"tempo"`
90
91 - `"verse"`
92
93 - `"vesper"`
94
95 - `"willow"`
96
97 - `CustomVoice object { id }`
98
99 - `id: string`
100
101 - `delegation: optional ClientDelegation or object { responses, type } or null`
102
103 Who handles tasks delegated by the Live model. Omitted or null selects your application; use `responses` to let the API manage a Responses backend.
104
105 - `ClientDelegation object { type }`
106
107 Delegate tasks to your application. The Live session emits delegation events that your backend handles.
108
109 - `type: "client"`
110
111 The delegation owner. Always `client` for tasks handled by your application.
112
113 - `"client"`
114
115 - `Responses object { responses, type }`
116
117 Delegate tasks to a Responses model managed by the Live session.
118
119 - `responses: ResponsesDelegationConfig`
120
121 Backend model, prompt, and tools used when the Live session delegates a task to Responses.
122
123 - `model: string`
124
125 The model used for server-owned Responses delegations.
126
127 - `instructions: optional string or null`
128
129 Instructions for the delegated Responses model, separate from Live instructions. See [backend prompting](/api/docs/guides/live-delegation#start-with-your-existing-backend-prompt).
130
131 - `max_output_tokens: optional number or null`
132
133 Maximum number of output tokens for each delegated response.
134
135 - `parallel_tool_calls: optional boolean or null`
136
137 Whether the delegated Responses model may request multiple tool calls in a single response.
138
139 - `reasoning: optional object { effort, summary } or null`
140
141 Reasoning settings passed to each delegated Responses request.
142
143 - `effort: optional "none" or "minimal" or "low" or 3 more or null`
144
145 How much reasoning effort the delegated Responses model should use. Supported values depend on the backend model.
146
147 - `"none"`
148
149 - `"minimal"`
150
151 - `"low"`
152
153 - `"medium"`
154
155 - `"high"`
156
157 - `"xhigh"`
158
159 - `summary: optional "concise" or "detailed" or "auto" or null`
160
161 The reasoning summary to request from the delegated Responses model, when supported.
162
163 - `"concise"`
164
165 - `"detailed"`
166
167 - `"auto"`
168
169 - `service_tier: optional "auto" or "default" or "fast_tier_temp_pilot" or 3 more or null`
170
171 Service tier for delegated Responses requests.
172
173 - `"auto"`
174
175 - `"default"`
176
177 - `"fast_tier_temp_pilot"`
178
179 - `"flex"`
180
181 - `"priority"`
182
183 - `"ultrafast"`
184
185 - `text: optional object { verbosity } or null`
186
187 Text generation settings passed to each delegated Responses request.
188
189 - `verbosity: optional "low" or "medium" or "high" or null`
190
191 The amount of detail in text generated by the Responses backend. This does not configure the Live model’s spoken delivery.
192
193 - `"low"`
194
195 - `"medium"`
196
197 - `"high"`
198
199 - `tool_choice: optional "auto" or "none" or "required" or object { name, type } or object { name, server_label, type }`
200
201 Controls which tool the Responses backend uses when handling a task delegated by the Live model.
202
203 - `LiveToolChoiceEnum = "auto" or "none" or "required"`
204
205 - `"auto"`
206
207 - `"none"`
208
209 - `"required"`
210
211 - `LiveFunctionToolChoiceParam object { name, type }`
212
213 - `name: string`
214
215 - `type: "function"`
216
217 - `"function"`
218
219 - `LiveMCPToolChoiceParam object { name, server_label, type }`
220
221 - `name: string`
222
223 - `server_label: string`
224
225 - `type: "mcp"`
226
227 - `"mcp"`
228
229 - `tools: optional array of FunctionTool or object { type }`
230
231 Tools available to the Responses backend while it handles tasks delegated by the Live model.
232
233 - `FunctionTool object { name, type, description, 2 more }`
234
235 A function tool available to the Responses backend when the Live model delegates a task.
236
237 - `name: string`
238
239 The name the delegated Responses model uses when calling this function.
240
241 - `type: "function"`
242
243 The tool type. Always `function`.
244
245 - `"function"`
246
247 - `description: optional string or null`
248
249 What the function does and when the delegated Responses model should call it.
250
251 - `parameters: optional map[unknown] or null`
252
253 A JSON Schema object describing the arguments accepted by the function.
254
255 - `strict: optional boolean or null`
256
257 Whether the delegated Responses model must follow the function’s parameter schema exactly.
258
259 - `WebSearch object { type }`
260
261 A web search tool available to the Live session’s Responses backend.
262
263 - `type: "web_search"`
264
265 The tool type. Always `web_search`.
266
267 - `"web_search"`
268
269 - `type: "responses"`
270
271 The delegation owner. Always `responses` for tasks handled by the Responses API.
272
273 - `"responses"`
274
275 - `input: optional array of InitialItem`
276
277 Ordered text-only history supplied before startup. Supports developer, user, and assistant messages with one text part each; at most 128 messages and 8,192 rendered tokens in total.
278
279 - `Developer object { content, role, id, 2 more }`
280
281 A developer message included in the initial text history of a Live session.
282
283 - `content: array of object { text, type }`
284
285 The message content. Supply exactly one text part for the initial Live conversation history.
286
287 - `text: string`
288
289 The message text to include in the Live session’s initial conversation history.
290
291 - `type: optional "input_text"`
292
293 The text content type. Always `input_text`.
294
295 - `"input_text"`
296
297 - `role: "developer"`
298
299 The author of this history message. Always `developer`.
300
301 - `"developer"`
302
303 - `id: optional string or null`
304
305 An optional identifier for the supplied history message. Live uses the message’s role and text to initialize the conversation.
306
307 - `status: optional "incomplete" or "completed" or null`
308
309 The supplied message’s status. Live uses its text as history and does not resume an incomplete message.
310
311 - `"incomplete"`
312
313 - `"completed"`
314
315 - `type: optional "message"`
316
317 The history item type. Always `message`.
318
319 - `"message"`
320
321 - `User object { content, role, id, 2 more }`
322
323 A user message included in the initial text history of a Live session.
324
325 - `content: array of object { text, type }`
326
327 The message content. Supply exactly one text part for the initial Live conversation history.
328
329 - `text: string`
330
331 The message text to include in the Live session’s initial conversation history.
332
333 - `type: optional "input_text"`
334
335 The text content type. Always `input_text`.
336
337 - `"input_text"`
338
339 - `role: "user"`
340
341 The author of this history message. Always `user`.
342
343 - `"user"`
344
345 - `id: optional string or null`
346
347 An optional identifier for the supplied history message. Live uses the message’s role and text to initialize the conversation.
348
349 - `status: optional "incomplete" or "completed" or null`
350
351 The supplied message’s status. Live uses its text as history and does not resume an incomplete message.
352
353 - `"incomplete"`
354
355 - `"completed"`
356
357 - `type: optional "message"`
358
359 The history item type. Always `message`.
360
361 - `"message"`
362
363 - `Assistant object { content, role, id, 2 more }`
364
365 An assistant message included in the initial text history of a Live session.
366
367 - `content: array of object { text, type } or object { text, type }`
368
369 The message content. Supply exactly one text part for the initial Live conversation history.
370
371 - `Text object { text, type }`
372
373 Assistant text supplied as conversation history when starting a Live session.
374
375 - `text: string`
376
377 The message text to include in the Live session’s initial conversation history.
378
379 - `type: optional "text"`
380
381 The text content type. Always `text`.
382
383 - `"text"`
384
385 - `OutputText object { text, type }`
386
387 Assistant output text supplied as conversation history when starting a Live session.
388
389 - `text: string`
390
391 The message text to include in the Live session’s initial conversation history.
392
393 - `type: "output_text"`
394
395 The text content type. Always `output_text`.
396
397 - `"output_text"`
398
399 - `role: "assistant"`
400
401 The author of this history message. Always `assistant`.
402
403 - `"assistant"`
404
405 - `id: optional string or null`
406
407 An optional identifier for the supplied history message. Live uses the message’s role and text to initialize the conversation.
408
409 - `status: optional "incomplete" or "completed" or null`
410
411 The supplied message’s status. Live uses its text as history and does not resume an incomplete message.
412
413 - `"incomplete"`
414
415 - `"completed"`
416
417 - `type: optional "message"`
418
419 The history item type. Always `message`.
420
421 - `"message"`
422
423 - `instructions: optional string or null`
424
425 Frontend instructions for voice, conversation, interruptions, and when to delegate. Start with the [Live prompting guide](/api/docs/guides/live-prompting); put business rules and tool workflows in a separate [backend prompt](/api/docs/guides/live-delegation#start-with-your-existing-backend-prompt). Limited to 16,384 client-supplied tokens. Omitted or blank instructions use server defaults. Immutable after startup.
426
427 - `store: optional boolean`
428
429 Whether to store the session for later forking and recording download. Defaults to false for new sessions.
430
431### Example
432
433```http
434curl https://api.openai.com/v1/live/sessions/$SESSION_ID/accept \
435 -H 'Content-Type: application/json' \
436 -H "Authorization: Bearer $OPENAI_API_KEY" \
437 -d '{
438 "session": {
439 "model": "gpt-live-1",
440 "type": "live"
441 }
442 }'
443```