REST: conversation.item.input_audio_transcription.completed
Written by
Updated at December 30, 2025
HTTP request
POST https://ai.api.cloud.yandex.net/v1/realtime/server-events/conversation.item.input_audio_transcription.completed
Body parameters
Request schema: application/json
{
"type": "string",
"event_id": "string",
"item_id": "string",
"content_index": "integer",
"transcript": "string",
"usage": "object"
}
Output of audio transcription for user audio written to the input audio buffer. Transcription begins when the buffer is committed (client or server VAD).
|
Field |
Description |
|
type |
enum Required field. The event type, must be
|
|
event_id |
string Required field. The unique ID of the server event. |
|
item_id |
string Required field. The ID of the item containing the audio that was transcribed. |
|
content_index |
integer Required field. The index of the content part in the item's content array. |
|
transcript |
string Required field. The transcribed text. |
|
usage |
object Token usage information (when provided by the backend). |
Response
HTTP Code: 200
OK