Translate media

Media translation is asynchronous. Register the source, start translation, then poll the reusable job.

1. Register media

Choose a path in Upload and register source files. Save the successful job_id.

2. Start translation

export LINGOPAL_API_KEY="your-api-key"
export JOB_ID="replace-with-job-id"

curl -X POST "https://api.lingopal.ai/v2/jobs/$JOB_ID/translations" \
  -H "X-API-Key: $LINGOPAL_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "target_languages": ["es", "fr"],
  "number_of_speakers": 0,
  "reset": true
}'

curl -X POST "https://example.com/v2/jobs/job_123/translations" \  -H "Content-Type: application/json" \  -d '{    "target_languages": [      "es",      "fr"    ]  }'

{  "workflow": "translate_media",  "status": "started",  "job_id": "job_123",  "target_languages": [    "es",    "fr"  ],  "message": "Translation workflow started",  "links": {    "job": "/v2/jobs/job_123",    "status": "/v2/jobs/job_123/status"  }}

Use the API Reference entry for translateJob for complete request and response behavior.

Media-specific options include background-audio removal, speaker-count hints, subtitle burn-in, AI-warning text, and enhancement. Use the API Reference for exact fields.

3. Monitor the workflow

Poll GET /v2/jobs/{job_id}/status until completed or failed. Wait for an active run to finish before submitting another translation for the same job.

See Job lifecycle for status fields, output flags, and rerun behavior.

4. Retrieve outputs

Download transcript and subtitle outputs through the public output endpoints. The current public contract does not expose a separate translated audio/video download endpoint; do not assume a media file URL unless it appears in your job response.