TexaData

Built to reduce your ingest cost, not add to it.

Most supply arrives as a hard drive of mixed formats with no metadata and no per-asset paperwork, and the buyer absorbs the cost of fixing it. We deliver in one format, one schema, one folder structure, deduplicated, with the rights already attached.

A welder at work on a kerb, a burst of sparks off the weld
The driver of a yellow bus at the wheel
A tailor working over a bolt of printed cloth
A market seen from above, dense with trading umbrellas
Ikoyi from the air, rooftops and towers running to the lagoon
A customer in the chair reflected in a barbershop mirror
01

Video

Container
MP4, H.264 high profile, faststart
Resolution
3840×2160 master; 1920×1080 proxy
Frame rate
30 fps constant; 60 fps for high-motion, declared per asset
Colour
Rec.709, 8-bit, SDR. No HDR, no Dolby Vision, no camera LUTs
Audio
AAC 160 kbps or PCM, 48 kHz
02

Audio

Format
WAV, 16-bit PCM, 48 kHz
Channels
Separated where the source allows; mix documented where not
Transcript
JSON, timestamped, speaker-attributed
Metadata
Language, accent, speaker count, overlap ratio, acoustic conditions
03

Metadata & rights

Manifest
CSV, 38 columns, one row per asset
Annotations
JSON, versioned schema, per clip
Rights
Scanned instruments named to asset ID
Payments
Log with a reference per contributor
Data card
Markdown, ships at the root of every delivery
04

Transfer

Methods
S3 or GCS bucket delivery, signed URLs, or physical media
Naming
Zero-padded asset IDs, no spaces, stable across versions
Review
Proxy pack sized for a single download
Masters
Released on execution
Traffic and roadside trading seen from above
Nigeria — capture environment · editorial

Manifest schema — all 38 columns

One row per deliverable asset. Technical columns are extracted from the container automatically at ingest; semantic columns are filled by the annotator who was present at capture.

  1. 01asset_id
  2. 02filename
  3. 03modality
  4. 04duration_sec
  5. 05resolution
  6. 06fps
  7. 07video_codec
  8. 08audio_channels
  9. 09audio_sample_rate
  10. 10file_size_bytes
  11. 11capture_date
  12. 12capture_time_local
  13. 13timezone
  14. 14country
  15. 15city
  16. 16location_type
  17. 17environment_tag
  18. 18camera_position
  19. 19camera_model
  20. 20subject_ids
  21. 21identifiable_persons
  22. 22languages
  23. 23primary_task
  24. 24intent
  25. 25objects_handled
  26. 26hands_visible
  27. 27failure_retry_events
  28. 28motion_level
  29. 29lighting
  30. 30consent_doc_ref
  31. 31contributor_licence_ref
  32. 32rights_holder
  33. 33exclusivity
  34. 34ai_training_grant
  35. 35sublicensable
  36. 36payment_ref
  37. 37payment_status
  38. 38notes