Deck: echo what voice awareness hears into the feed - #22
Merged
Conversation
Every transcribed line (streamer mic, and the party channel when enabled) now shows in the deck feed as a faint italic row - mic lines right-aligned like your typed messages, party lines left with their channel label. When a ghost reacts strangely you can see exactly what the transcription thought you said. Deliberately sensor-readout, not chat: heard lines never enter the history the brain sees (it already gets the transcript window), never persist, never reach the recap or the on-stream overlay, and the deck caps them at 30 rows. transcript.showInFeed (default on, Settings -> Voice) hot-applies without a restart. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01FNRQP2h3GUQVThuE2KVBEu
The heard-in-feed echo (and voice replies) hang off TranscriptFeed's onSpeech callback, which both modes share - lock that in with a textSource-mode test covering priming, unchanged-caption dedupe, OBS unreachable, and SRT-metadata filtering. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01FNRQP2h3GUQVThuE2KVBEu
MyNamesEMurray
enabled auto-merge
July 18, 2026 00:28
# Conflicts: # src/server.js
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Every line voice awareness transcribes — your mic, and the party channel if enabled — now appears in the deck feed as a faint 🎙 row, so when a ghost reacts oddly you can see exactly what the transcription thought you said (mishears included).
Changes
heardWS event (channel: mic|party, party carries its configured label). Gated bytranscript.showInFeed(default on), read live and registered as a hot path — toggling it in Settings applies without a relaunch.history— the brain already receives the transcript window separately, so no double-feeding or prompt pollution;config.example.jsondefault, README config row + voice-awareness bullet.No
Release-Bumptrailer → rides as a patch. Shout if you'd rather this one counted as a minor.Verification
heard{channel:mic}, party line asheard{channel:party, label:"Discord"}; togglingtranscript.showInFeedoff saves withapplied: true(hot, no restart) and suppresses subsequent events whilestate.lastHeardkeeps updating; chathistorystays empty throughout.Risk & rollout
Checklist
npm run smokepasses locally (or CI is green)main🤖 Generated with Claude Code
https://claude.ai/code/session_01FNRQP2h3GUQVThuE2KVBEu
Generated by Claude Code