Introducing GstRTMPMetadata: Unlocking Metadata in RTMP Streams
- oscarporras7
- Oct 27
- 2 min read
Live streaming today isn’t only about video and audio. Applications in AI, robotics, and broadcasting demand contextual information - detections, telemetry, captions, and more - that travels alongside the stream.
While RTMP reliably carries audio, video, and standard metadata into platforms like YouTube, Twitch, and Facebook Live, it wasn’t designed to handle custom, application-specific metadata. Developers who need to send analytics, sensor data, or AI events typically build side channels or custom APIs, adding unnecessary complexity and increasing development time.
Why RidgeRun Built GstRTMPMetadata
To solve that, RidgeRun developed GstRTMPMetadata, a GStreamer plugin that brings native metadata support to RTMP. No side channels. No extra APIs. Just one clean pipeline that carries both video and data.
With GstRTMPMetadata, you can enrich your live video with contextual information such as AI detections, telemetry, or captions, directly within the RTMP stream.
How It Works
Sending Metadata (flvmux)
GstRTMPMetadata enhances the flvmux element to let you embed metadata using several flexible methods:
meta-string – Add metadata as string (e.g., "Hello RTMP Metadata"). Ideal for debugging or simple integrations.
meta-binary – Inject structured binary data, useful for sending JSON, protobufs, or sensor packets.
GStreamer GstMeta injection – Upstream elements can attach custom GstFlvMeta objects to buffers, which are automatically serialized into the RTMP stream.

Receiving Metadata (flvdemux)
On the receiving end, the modified flvdemux element can extract metadata from the same RTMP stream.
Applications can then use the metadata independently without interfering with video playback.

By making metadata a native part of the streaming pipeline, RidgeRun enables more integrated and flexible workflows for live streaming and video analytics applications.
How This Impacts Your Product
With RidgeRun’s GstRTMPMetadata, you can:
Streamline your workflow — one pipeline for video and metadata.
Reduce integration time — adding metadata is just a property away.
Scale easily — works across major GStreamer releases and embedded platforms like NVIDIA Jetson, NXP i.MX8, and Qualcomm.
Deliver richer experiences — combine live video with AI insights, telemetry, or captions.
Where It Fits Best
Live broadcasting: add real-time stats, overlays, or ad triggers.
Robotics & drones: send telemetry and AI detections together with the video feed.
Medical streaming: transmit vital signs with surgical video.
Industrial inspection: enrich streams with sensor and measurement data.
Try It Yourself
Experience how much complexity disappears when metadata becomes part of your RTMP workflow. Ask for GstRTMPMetadata evaluation today and explore examples on our GstRTMPMetadata Wiki



