Conversation
 |
Independent Volume was the only code path that wrote the macOS output volume. It saved the current level, set the output to the selected cue level for the length of the cue, then restored it, which is what let a transcription cue change other applications' audio (altic-dev#522). Compensating app-side instead does not preserve what the option promised. The CoreAudio volume scalar and AVAudioPlayer.volume are different gain domains, so a player gain derived from their ratio does not land on the selected level, and it caps at full gain once the selected level is above the current output. Removing the option leaves one cue-volume slider, and nothing writes the system volume any more. It also removes four failure modes that lived in the save/restore path: a failed CoreAudio read was saved as 1.0 and restored as maximum, the restore re-resolved the default output so switching outputs mid-cue wrote one device's baseline onto another, the delayed restore had no termination handler, and overlapping cues shared one saved-volume slot. The Independent Volume settings copy is removed with the toggle. It described the old behaviour ("stays constant regardless of system volume", "temporarily changes system volume during playback") and was wrong in both directions either way.
FluidVoice PR build readyDownload FluidVoice-PR-989-db4e09d97e74 The artifact contains the ad-hoc-signed app ZIP, Xcode archive, build manifest, and installation instructions. It expires 5 days after the build. Install the app
This build has its own app identity, so its permissions are separate from the release version of FluidVoice. |
…orts Dropping transcriptionSoundIndependentVolume from SettingsBackupPayload broke downgrades: the previous app version's synthesized decoder still requires that key, and the schema stays 1.0, so a backup written by this build looks compatible and is reported as invalid JSON instead. Encode the key again with a fixed false and decode it tolerantly. Nothing reads it back into settings, so Independent Volume stays removed.
b7dd082 to
db4e09d
Compare
1 participant
Description
This is the change you floated on #867, dropping the Independent Volume option outright instead of keeping it with app-side compensation. I pushed it on 2026-09-02, and the stale bot closed #867 on 2026-09-18 for inactivity, so this is the same branch at the same head (b7dd082) as a fresh PR. Ignore my 2026-09-10 comment on that thread saying the revision was still coming. It was already in at that point.
The toggle is gone, and so is the save, set and restore code that wrote the Mac output volume. What is left is the one cue-volume slider. The cue plays at that level through
AVAudioPlayer, so it scales with system output like any other sound, and nothing in the app writes the system volume any more.One thing stays on purpose. The backup payload still encodes
transcriptionSoundIndependentVolumeasfalse, because the schema is still 1.0 and the previous version's decoder requires the key. Nothing reads it back into settings, so the setting itself is gone.Type of Change
Related Issue or Discussion
Closes #522. Replaces #867, which the stale bot closed. This removes the only place the app wrote the system volume; if the spike in #522 still reproduces with the toggle gone, that is a different write and worth reopening.
Testing
swiftlint --strict --config .swiftlint.ymlswiftformat --config .swiftformat SourcesSame head as #867, so the checks there still hold. CI was green on b7dd082: Build FluidVoice, SwiftLint and the PR archive all passed. When I pushed it on #867,
swiftlint --strictover the whole repository reported 7 violations, alllegacy_swiftui_aspect_ratioand all in files this branch does not touch; a clean checkout ofmainreported the identical 7.Locally I ran
xcodebuild build-for-testingrather than the full suite at that time, so the test target compiles and CI ran the tests. The eight gain-policy tests this branch added earlier are gone, because the policy they covered no longer exists.DictationE2ETests.swiftcarries one test instead: encode a backup, check the deprecated key is written asfalse, strip the key, check the payload still decodes.It still merges clean into current
main, so no rebase.Screenshots / Video
Settings sound section on current main. This PR removes the Independent Volume row.