Skip to main content
GET
Get Vision

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Path Parameters

project_id
string
required

Response

Successful Response

project_id
string
required
status
enum<string>
required

Analysis lifecycle as reported by the result GET endpoints.

Available options:
queued,
processing,
completed,
failed
analysis
string
default:vision
Allowed value: "vision"
error
string | null
agent_context
Agent Context · object
faces
Faces · object[]
face_tracks
Face Tracks · object[]
scenes
Scenes · object[]
asd
VisionAsdStatus · object | null

Whether active-speaker detection ran for this job, and what it produced.

Detections that were never scored omit speak_conf entirely, so this record is the only way to tell "never ran" from "ran and scored nothing". Active- speaker detection needs at least two simultaneously tracked faces, so single-speaker footage produces no scores by design.