Skip to main content

Voice Briefing Delivery

Learning goals​

Demonstrate parallel public output, browser rendering and an external side effect.

Quick reference: Level 5; advanced; output type briefing_text; estimated cost: LLM, speech and storage provider dependent.

Architecture​

Text alternative: The graph contains FlowBeads/input, LLMBeads/LLMProcess, Speech/ElevenLabsTTS, Outputs/Audioplayer, Outputs/S3WriteFile, FlowBeads/output. It accepts public input and returns a public result.

Prerequisites​

None.

Connections​

  • Configured S3 destination

Secrets​

  • One supported LLM secret
  • ElevenLabs secret
  • S3 credentials

Build and run​

  1. Configure all three providers.
  2. Set the S3 destination.
  3. Run and inspect text, audio and storage events.

Use this fixture in the Test panel:

Briefing request
Create a concise morning briefing about the supplied topic.

Expected result​

  • text is returned
  • audio sink is available
  • S3 write event is audited

Open the persisted run and verify the node timeline. Browser sinks appear in the Inspector; public output appears in the run result when enabled.

Side effects and cost​

  • Writes one audio object to S3

Cost: LLM, speech and storage provider dependent. DataBraid stores summaries and structured errors, not secret values.

Extensions​

  • Change the fixture and compare the node-level timeline.
  • Add a Data Probe before the final output.
  • Save a known-good fixture for regression testing.