Media solutions
Build calling and media-server applications, add AI processing, or create production and editing tools. CoreMF supplies the media foundation; your team builds the product experience and operations.
Need a particular function instead? Search all capabilities ↗ · Compare build vs. buy ↗
Real-time communication
Build private calling, meeting or interactive media products with shared media and recording. Reuse CoreMF RTC/SFU paths and browser reference clients with chat, whiteboard and camera background effects.
Participants / shared media → RTC / SFU → viewers + recording
Server MCU composition is a development/reference path; its CPU/GPU menu entries are hidden in release packages.
Intelligent media
Add detection, tracking or custom analysis through native, Python or JavaScript plugins with video frame and audio sample access. Connect processed media and timestamped results to overlays, events or downstream application logic.
Media → processor / model → events, overlays + downstream media
Explore plugin interfaces ↗Plugin interfaces and runtime costs differ. Processors/models are selected separately; no default AI model pack.
Broadcast, webinars & production tools
Create broadcast and webinar production tools with GPU layouts, titles, transitions and custom shaders. Reuse templates and preview/program control; drive graphics from timestamped AI or application metadata.
Sources + graphics → CoreGL GPU processing → program / record / stream
Watch actual CoreGL output ↗Media gateways & live backends
Connect cameras, encoders and remote feeds to browser audiences, recordings or streaming outputs. Build RTMP/RTSP/SRT/RTP ingest and processing branches on reusable graphs, with RTC/SFU or segmented-stream delivery.
Camera / encoder / network → ingest / process → record + publish
Media services & content delivery
Add file conversion or adaptive streaming to your application or service. Reuse codec/container conversion and HLS/DASH output, with automatic encoder selection, hardware encoding or explicitly selected customer BYO encoding.
File / live input → convert / encode → MP4, HLS or DASH
runAPI -i input.mp4 -o output.mp4 -p tc -s 1280x720 -r 30
runAPI -i input.mp4 -o master.m3u8 -p hls -n 3Editing, authoring & render services
Create editing tools or render services using timelines, seek/scrub preview, transactions and render jobs. Build your own interface and editing rules on the production engine.
Assets + timeline → plan / preview / render → media output
Requires the production/editing profile. CoreStudio is not included.
Devices, desktop tools & appliances
Create desktop tools or appliances for capture, playback and recording. Connect cameras, microphones and screen/window capture to local previews or network output, including processed GPU video.
Device / screen / stream → process → preview / record / publish
Device backends and GPU presentation are platform/build-specific.
Voice, broadcast & audio applications
Add selected mixing, effects and monitoring to voice or broadcast applications alongside video. CoreAudio is still maturing; validate the effects and synchronization required by your delivery.
Audio inputs → mix / effects / analysis → encode / play / metadata
runAPI -i input.mp4 -af gain -audio-filter-config '{"gain_db":-6}' -o adjusted.mp4 -p tcPlayers, review tools & content navigation
Add visual seek previews to players and media review tools. Generate sprite sheets and WebVTT navigation metadata with configurable sampling and layout.
Video → sampled frames → sprite sheets + WebVTT
runAPI -i input.mp4 -o storyboards -p storyboardProtected media delivery
Build protected streaming or recording products using StreamCrypto routes or DRM sample encryption and HLS/DASH packaging. Integrate your key, entitlement and license authority with the media path. See the protection levels ↗
Media + key provider → protection / packaging → authorized player
Production DRM depends on customer key/license services and player/provider qualification.
Solution · customer-operated media infrastructure
Build a gateway from encoder feeds, files or capture devices to RTC rooms and recording. Add composition or segmented delivery when required.
Ingest → optional decode / filter / encode → SFU distribution
Selected streams → MCU composition + audio mix → SFU program /
recording
Forward compatible encoded publisher tracks to multiple viewers without decoding/re-encoding for every viewer. Normalize or transcode an ingest upstream when the target RTC codec/format requires it.
Decode selected sources, synchronize, compose video with CoreGL or CPU filters, mix audio and encode a program for publication or recording. The application controls source/room mapping and layout.
Each ingest example publishes one source to an SFU. A configured signaling service, compatible encoder/runtime and reachable endpoints are required; replace SIGNAL_HOST. Multiple ingest graphs and room bindings are assembled through the application API or Graph JSON.
runAPI -i "rtmp://0.0.0.0:1935/live/cam01" -o "wss://SIGNAL_HOST:8443/?id=cam01" -p sfupub
runAPI -i "rtsp://0.0.0.0:8554/live/cam02" -rtsp-tcp -o "wss://SIGNAL_HOST:8443/?id=cam02" -p sfupub
runAPI -i "srt://0.0.0.0:9000?mode=listener" -o "wss://SIGNAL_HOST:8443/?id=cam03" -p sfupubA separate meeting controller can compose publishers already mapped into its room. This is the development/reference server-MCU path:
runAPI -p meeting=studio -o "wss://SIGNAL_HOST:8443/?id=studio" --server-mcu --mcu-publishers auto --mcu-renderer gpu --record-mcu studio.mp4Room/job sharding, load balancing and autoscaling belong to the deployment. Capacity depends on codecs, resolution, composition, hardware and egress; measure the target workload. Server MCU remains development/reference scope, and current server packages target Linux. Release boundaries ↗
Connect these capabilities through CoreMF_API.h and Graph JSON. Agree native component/source access separately; the delivered build defines supported features.