data (video) → transform → stream
A reference technical solution for composing per-user HLS feeds from indexed moments. Retrieve clips by query, compose them on a chunk-level timeline, apply branding and subtitles as overlay tracks, publish as an HLS manifest. No re-encode step.
Request resolves against the indexes already built on your collection. Selected moments are composed on a Timeline; overlays apply at the manifest layer; output is an HLS m3u8 URL with per-segment caching.
Each stage is a method on the SDK. No infrastructure to operate. No re-encoding pipelines to run.
Query the indexes you've already built. Filters, semantic recall, and structured constraints in one call. Returns ranked moments with playable clip URLs.
# search moments results = video.search( "goals by player_42", type="deepsearch", ) shots = results.get_shots()
Merge the moments. Lay overlays on top. Add multilingual subtitles. Reframe for the target aspect ratio. All chunk-level operations. No re-encode.
# compose timeline = Timeline(conn) timeline.resolution = "608x1080" video_track.add_clip(0, highlight_clip) brand_track.add_clip(0, logo_clip) caption_track.add_clip(0, caption_clip)
Output is an HLS stream URL. Embeddable in any player, cacheable at the CDN edge, addressable per-user. Generated fresh, on demand.
# publish timeline.add_track(track) stream_url = timeline.generate_stream()
from videodb import connect from videodb.editor import Timeline, Track, Clip, VideoAsset, ImageAsset, CaptionAsset conn = connect() coll = conn.get_collection() video = coll.get_video("match_final") logo = coll.upload(file_path="brand.png") # 1. Retrieve. Query the indexes you already built. results = video.search("goals scored by player_42", type="deepsearch") moment = results.get_shots()[0] # 2. Compose. Add branding, captions, and mobile framing. timeline = Timeline(conn) timeline.resolution = "608x1080" # vertical for mobile video_track, brand_track, caption_track = Track(), Track(), Track() video_track.add_clip(0, Clip(asset=VideoAsset(id=video.id, start=moment.start), duration=12)) brand_track.add_clip(0, Clip(asset=ImageAsset(id=logo.id), duration=12, opacity=0.85)) caption_track.add_clip(0, Clip(asset=CaptionAsset(src="auto"), duration=12)) # 3. Stream. Fresh HLS URL, cacheable per-user. timeline.add_track(video_track) timeline.add_track(brand_track) timeline.add_track(caption_track) stream_url = timeline.generate_stream() print(stream_url) # → https://cdn.videodb.io/u/42/highlights.m3u8
The whole pipeline executes inside VideoDB. No round-trip to your services. No re-encoding step.
<1 s for first segment. End-to-end budget.
Retrieve is ~120 ms p95. Compose is chunk-pointer math. No re-encode. First HLS segment is served while later segments resolve in the background.
Each personalized manifest gets a unique URL; CDN segments shared across users are deduplicated. Per-user TTL keeps invalidation cheap.
Brand marks, subtitles, and reframes are applied as overlay tracks on the HLS manifest. Same source bits; new visual surface.
Pay for retrievals + unique segments served. No charge for compose-time CPU. The format makes the cuts free.
Per-fan highlights composed from live games and seasons of archive. Crowd-peak audio detection, player tracking, scoreboard overlay rendering at the manifest layer.
"Previously on" cuts generated per viewer based on the last episode watched and which character arcs they're following. Composed inline; output is a per-user HLS manifest.
Slot ads at scene boundaries, not on a fixed clock. Brand-safety and audience filters apply at compose time so the same source produces different manifests per viewer.
An agent pulls source clips from a knowledge base, composes them with synchronized subtitles, and streams the result back to the user. The agent owns the query; VideoDB owns the render.
