Live broadcasting
How newsrooms and live event broadcasters extend translated captions to global audiences without doubling production cost.
A newsroom covers a press conference in a language most of its audience does not speak. The traditional solution: cut away to a studio anchor who summarises what was said, supplemented by clips with voice-over translation. The audience gets the gist, but the moment — the speaker’s tone, the rhythm of the exchange, the unscripted follow-up — is lost in the summarisation.
Loquira offers a different solution: broadcast the original feed with translated captions overlaid. The audience watches the original event. They read the translation in their language, timed to the speaker’s delivery. The editorial narrative is the event itself, not the studio’s summary. For a deeper look at the distinction between these approaches, see live captions vs live translation.
Where Loquira fits in a news broadcast pipeline
A typical news broadcast chain for live events looks like:
Live feed → Ingestion → Graphics overlay → Playout → DistributionLoquira provides translated text and audio in real time. The listener view displays the translated transcript, which can be used as a reference for manual caption entry, or exported as SRT or WebVTT after the session for caption overlay on recorded content. See our guide on embedding live captions for integration details. Direct broadcast graphics integration (browser source, WebSocket feed) is on the product roadmap.
Latency budgets
Broadcasters operate within strict timing constraints. Every frame has a scheduled position. Adding translation cannot disrupt that schedule.
Loquira’s latency budget from microphone to translated output:
| Stage | Latency |
|---|---|
| Speech recognition (Deepgram Nova-3, paid tiers) | ~300 ms |
| Translation (Gemini Flash) | ~250 ms |
| Text-to-speech synthesis | ~200 ms |
| Total end-to-end | ~750 ms |
This is well within the delay buffer most broadcasters maintain for live events (typically 3–10 seconds for legal review and profanity delay). A live-to-air route is realistic for news coverage, press conferences, and live event broadcasting. For programming where every second counts — breaking news, live sports — the latency is imperceptible to the viewer.
Editorial oversight
For newsrooms that require editorial control over on-air captions, a caption operator monitors the Loquira listener view and enters translated text into the graphics system manually. This mirrors how captions are managed in traditional broadcast workflows: the operator is a production role, responsible for timing and visibility, not translation accuracy. The translation accuracy is handled by the engine and verified against the source transcript after the event.
Audience growth in non-working-language regions
The business case for translated broadcasting is audience reach. A news channel that covers a French-language press conference with English captions serves an audience that the original-language broadcast would not reach. The captions are the product — they make the content accessible to viewers who do not speak the source language.
After the event, the transcript — available in every session language via transcripts and exports — can be published as a companion article on the newsroom’s website. The article includes the full translated transcript in every language the broadcast supported, alongside the video replay with captions. This extends the shelf life of the content from the live broadcast window to the archive lifetime of the website.
For international broadcasters — organisations like France 24, DW, or Al Jazeera that already operate in multiple languages — Loquira allows a single production team to produce content for all language channels simultaneously. The press conference is covered once; each language channel receives its own caption feed and transcript without additional production overhead.