Skip to main content
POST
Start media analysis

Authorizations

X-License-Key
string
header
required

Jumper Pro license key passed via header

Body

application/json
cache_dir
string
required

Where to store analysis results

detect_speakers
boolean
default:true

Set to false to skip final speaker diarization for transcription jobs.

speaker_count
default:auto

Optional speaker-count hint for pyannote diarization. Use auto for unconstrained detection, 1-4 for num_speakers, or 5+ for min_speakers=5 and max_speakers=12.

Available options:
auto,
auto-detect,
5+
preferred_qwen_model_key
enum<string>
default:auto

Optional Qwen ASR preference for Qwen-supported languages. Use faster for the 0.6B model, more accurate for the 1.7B model, or auto/default to use Jumper's default.

Available options:
auto,
faster,
more accurate,
qwen3-asr-0.6b,
qwen3-asr-1.7b
visual_media_paths
string[]

Files to run visual analysis on

transcription_jobs
object[]

Files to transcribe, with optional language and channel hints

face_clustering_jobs
object[]

Face detection and clustering job definitions

Response

Analysis started

message
string
Example:

"Analysis started"

task_id
string

Use this ID to join a SocketIO room for progress events

Last modified on July 14, 2026