rbrooks/Quest-Board

Watch

v3.11.5 - Recording Clock Hotfix

New issue

P0. Emergency fix for the root cause found in #319: per-speaker audio tracks have never shared a session clock, so every multi-speaker recording ever produced has a scrambled timeline. Small, self-contained, testable in CI, and it must ship with a silence strategy or it is a regression. See docs/.internal/session-lifecycle-review-2026-08.md section 7.

2026-08-25

100% Completed

[Backend] GET /api/sessions/{id} silently drops five declared response fields

#417 by claude-bot was closed 2026-08-25 22:13:57 +00:00 0 / 5

[Backend] Mark sessions recorded before the timeline fix as unreliable

#328 by claude-bot was closed 2026-08-29 07:57:57 +00:00 0 / 5

[Recording] Log a per-speaker captured/expected ratio so a timeline regression is one grep away

#327 by claude-bot was closed 2026-08-29 07:57:56 +00:00 0 / 5

[Recording] Replace the capture tests that assert the broken clock model

#326 by claude-bot was closed 2026-08-29 07:57:54 +00:00 0 / 5

[Backend] Stop writing transcript text to logs, and scrub retained production logs

#325 by claude-bot was closed 2026-08-29 07:57:53 +00:00 0 / 5

[Backend] Fail processing when per-speaker track durations do not match the recorded session length

#324 by claude-bot was closed 2026-08-29 07:57:52 +00:00 0 / 5

[Backend] Enable VAD by default and drop the noise floor to -45 dB

#323 by claude-bot was closed 2026-08-29 07:57:50 +00:00 0 / 6

[Recording] Reset a speaker's Opus decoder when their SSRC changes

#322 by claude-bot was closed 2026-08-29 07:57:49 +00:00 0 / 4

[Recording] Move raw capture off the WAV container to headerless s16le

#321 by claude-bot was closed 2026-08-29 07:57:48 +00:00 0 / 5

[Recording] Anchor per-speaker capture to a wall clock — tracks have never shared a session timeline

#320 by claude-bot was closed 2026-08-25 23:11:51 +00:00 0 / 7