This article explains how to use InterScribe features effectively in your events.
InterScribe supports several industry-standard streaming protocols that allow you to ingest high-quality audio (and optional video) from professional production tools like OBS, vMix, Wirecast, or hardware encoders. These protocols are ideal when you're already producing a livestream and want to generate real-time captions and translations for remote audiences.
⚠️ Important: These protocols are not recommended for in-person or interpreter-facing events due to the inherent delay. For low-latency use cases, see Desktop Agent Setup.
Protocol | Latency | Compatibility | Ideal Use Case | Notes |
---|---|---|---|---|
RTMP | 5–10 sec | Widely supported | Online-only | Simple setup, but high latency |
SRT | 2–5 sec | Modern encoders | Online-only (Remote production with unstable networks) | More complex setup |
WHIP | ~1 sec | Emerging tools | In-Person, Hybrid (Experimental low-latency streams) | Cutting-edge, limited support |
Overview: RTMP is the most common protocol for streaming video and audio. It’s supported by virtually all livestreaming tools and platforms, including OBS, Wirecast, vMix, and many hardware encoders.
✅ Pros:
⚠️ Cons:
💡 Best For: Pre-recorded streams, keynote broadcasts, or remote-only events where delay is acceptable.
Overview: SRT is a secure, open-source protocol built for low-latency and resilient transmission over unstable or unpredictable networks.
✅ Pros:
⚠️ Cons:
💡 Best For:
Overview: WHIP is a new WebRTC-based protocol designed to enable ultra-low-latency streaming to media servers over a simplified HTTP interface.
✅ Pros:
⚠️ Cons:
💡 Best For: In-Person sessions
While this page focuses on external protocols like RTMP, SRT, and WHIP, it's important to note that InterScribe’s own audio input tools — including the Streamer Dashboard, Desktop Agent, and Web Agent — use WebRTC-based streaming behind the scenes.
Benefits of InterScribe’s built-in WebRTC streaming:
✅ Recommended: Use these agents for sessions where real-time responsiveness is critical — such as interpretation booths, hybrid meetings, or local audiences.
If you're using external encoders (like OBS or vMix), then RTMP, SRT, or WHIP are your available protocols — but expect added latency due to the nature of those formats.
🟢 Use InterScribe Native Agents (WebRTC) — including the Streamer Dashboard, Desktop Agent, and Web Agent — for the lowest latency and best experience in real-time and in-person events.
🟡 Use RTMP for maximum compatibility with tools like OBS, vMix, or hardware encoders — but only for online-only events, as it introduces noticeable delay.
🟠 Choose SRT when you need more resilience over unstable networks. It offers better performance than RTMP, but is also best suited for online events due to moderate delay.
🔴 Try WHIP only if you're working with a compatible encoder or integration and can’t use InterScribe's native agents. WHIP is low-latency but requires advanced configuration and is still emerging.
✅ Always test your full setup — including the audio source, ingress configuration, and session playback — before going live.
🎯 Make sure you've created the required Ingress under A/V Inputs → Ingresses and assigned it to the correct AV Channel before streaming.
You can create multiple Ingresses for different AV Channels, each using a different protocol. However, each session can only receive audio from one assigned AV Channel at a time.
Yes — all protocols support video, but InterScribe only uses the audio for captioning and interpretation. You can embed the video (e.g. YouTube/Vimeo) separately in the session.
No. OBS/vMix typically use RTMP or SRT. WebRTC is used internally by InterScribe's own agents like the Streamer Dashboard or Desktop Agent.
WebRTC via the Desktop Agent or Streamer Dashboard is ideal due to ultra-low latency and ease of configuration.
No — these protocols are for external encoders only. The Desktop Agent and Web Agent use direct browser or system-level audio streaming instead.
Would you like a visual diagram comparing all input methods and protocols across use cases? I can generate one.
We respect your privacy. We respect your privacy.
TLDR: We use cookies for language selection, theme, and analytics. Learn more. TLDR: We use cookies for language selection, theme, and analytics. Learn more