Commit Graph
6 Commits
Author SHA1 Message Date
Sameh Khamis afb3a92914 docs: say what the connection reads from the voice credential
It reads the token, model, transcription model and tier for diagnostics and
cost accounting, so the credential is not opaque to it. Also lists requestToken's
authMode option in APPLICATION.md.
2026-10-14 11:59:51 -07:00
Sameh Khamis af95f957ec Merge branch 'main' into docs/voice-backend-contract 2026-10-14 11:59:03 -07:00
Bilawal Sidhu a19897ad15 feat(voice): voice core — honest layer reach, spoken summary with on-screen card, narration, point-and-ask
- Voice layer manifest (src/voice/layerManifest.js) generates the layer enums,
  aliases and analyst field hints; a test fails when a catalog layer has no entry.
- analyst_query validates fields, operators, units, values and scopes, counts up
  to 250,000 loaded records per layer, reports complete/partial/unanswered, tells
  LAYER_OFF / NOT_READY / FEED_UNAVAILABLE apart from zero, refuses mismatched
  follow-ups and never lets a superseded query replace follow-up memory. Seven
  more layers expose analyst records.
- Tool results carry a code-built say line and a display card; the voice card in
  the mic control shows captions, plan steps and the result
  (src/voice/resultDisplay.js is the one adapter). Slow tools narrate their
  current step out of band; holding Space while the assistant talks cuts it off.
  Turn spans go to the debug log; caption transcription is metered; the log
  omits what was said unless GEV_VOICE_LOG_CONTENT=1.
- Point and ask: the Space-down pointer snapshot resolves "this", "that one" and
  "around here" through 'pointer' targets, and "the second one" through
  referent:n from the last numbered list.
- Prompt states each policy once (instructions + tools 48.5k -> 46.7k chars).
- Voice benchmark harness (scripts/qa-voice-bench.mjs, scripts/voice-bench/)
  shares its phrase table and grading with scripts/qa-voice-routing.mjs.
2026-09-28 21:19:23 -05:00
daikaginza 0da4d390ea docs: add the realtime backend to the voice ownership table
The ownership table lists every owner except the backend, though realtimeBackend.js owns the two requests, their separate transports, the abort/timeout scope around both, and client-secret expiry.

The method signatures are already documented with application construction, so link there rather than repeat them, and state only the ownership facts written nowhere: the connection is the backend's only consumer in the application, the resolved credential is opaque to it and is handed to negotiate whole, and client-secret expiry is refused inside the backend.
2026-09-26 17:36:59 +07:00
Sameh Khamis e7970c50b7 fix(voice): discard continuations from closed conversations 2026-09-15 12:13:35 -07:00
Sameh Khamis cc9a080f2d refactor(voice): separate Realtime state and resource owners 2026-09-15 12:05:01 -07:00