Other Changes
- (bug) Stop dictation from hanging on "Preparing microphone" after a failed microphone switch (#1979)
- (bug) Stop reporting a missed text field in agterm and let users add terminal apps (#1977)
- (feature) Label subtitle exports as subtitles with timecodes (#1976)
- (bug) Judge the spoken language from the first speech, not the first loud sound (#1972)
- (bug) Tell the two people of short recordings apart in speaker detection (#1969)
- (bug) Let workflows saved with a renamed transcription model be saved again (#1968)
- (feature) Move the AssemblyAI plugin to AssemblyAI-STT-3.7 and offer two live models (#1967)
- (bug) Keep AirPods out of call-quality audio while TypeWhisper is idle (#1962)