blitztext-app-linux/linux/blitztext
mARTin-B78 5aafb572ed feat: dedicated wakeword models for cancel and send actions (v2.03.30)
WakewordActionListener opens a second Wyoming connection during active
wakeword recording, listening for the configured cancel/send models. When
either fires it immediately calls cancel_dictation() or finish_dictation()
without any silence timer or Whisper pass. Settings UI adds Cancel model
and Send model pickers to the wakeword config card, populated from the
same server model list as the trigger model.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-10 12:07:03 +02:00
..
__init__.py feat: dedicated wakeword models for cancel and send actions (v2.03.30) 2026-06-10 12:07:03 +02:00
__main__.py Add STT + LLM engine presets, rename entry to blitztext.py, macOS app icon 2026-06-04 20:21:03 +02:00
audio.py feat: real-time cancel keyword detection during wakeword recording (v2.03.29) 2026-06-10 11:47:10 +02:00
autostart.py gtksettings: align Input tab labels to uniform 175px width; autostart: add 12s delay 2026-06-08 20:00:33 +02:00
benchmark.py feat: server RAM via Prometheus /metrics in benchmark (v2.03.09) 2026-06-09 23:24:40 +02:00
blitztext.py Stop notification storms: transient notifications, single instance, no empty-result alerts 2026-06-06 10:42:13 +02:00
caret.py stability + settings: fix AT-SPI session freeze, markdown Manual, bigger emojis (v1.9.6) 2026-06-09 14:20:05 +02:00
config.py feat: dedicated wakeword models for cancel and send actions (v2.03.30) 2026-06-10 12:07:03 +02:00
daemon.py feat: dedicated wakeword models for cancel and send actions (v2.03.30) 2026-06-10 12:07:03 +02:00
gtksettings.py feat: dedicated wakeword models for cancel and send actions (v2.03.30) 2026-06-10 12:07:03 +02:00
gtkui.py feat: overlay × cancel button + wakeword shortcut fields inline (v2.03.27) 2026-06-10 04:00:23 +02:00
inputmode.py fix: cancel recording works in wakeword mode + tray Cancel item (v2.03.26) 2026-06-10 03:44:14 +02:00
llm.py llm: catch all exceptions, validate response, show transcription preview (v1.9.9) 2026-06-09 15:02:45 +02:00
logbuffer.py fix: stability, WhisperX 404, log levels (v2.03.00) 2026-06-09 22:01:14 +02:00
notify.py Stop notification storms: transient notifications, single instance, no empty-result alerts 2026-06-06 10:42:13 +02:00
overlay.py feat: overlay × cancel button + wakeword shortcut fields inline (v2.03.27) 2026-06-10 04:00:23 +02:00
paste.py fix: auto-paste long/multiline output, O(n²) overlay accumulation (v2.0.1) 2026-06-09 15:56:44 +02:00
quality.py dictation: spoken newlines + LLM thinking animation (v1.9.8) 2026-06-09 14:43:15 +02:00
recorder.py Rewrite README; fix fd leak, StopIteration crash, add pip Dependabot 2026-06-05 16:01:31 +02:00
routing.py routing: send-by-voice keyword (types + presses Enter); wakeword benchmark via TTS; Release 1.8.0 2026-06-08 13:17:33 +02:00
sound.py feat: MP3/OGG/FLAC sound support + browse dialog with auto-preview (v2.03.16) 2026-06-10 00:37:07 +02:00
streaming.py Add realtime streaming and Linux app docs 2026-06-05 15:06:17 +02:00
stt.py fix: probe Prometheus /metrics at server root, not under /v1 (v2.03.10) 2026-06-09 23:30:33 +02:00
transcribe.py feat: real-time cancel keyword detection during wakeword recording (v2.03.29) 2026-06-10 11:47:10 +02:00
tray.py fix: cancel recording works in wakeword mode + tray Cancel item (v2.03.26) 2026-06-10 03:44:14 +02:00
wakeword_bench.py fix: suppress numpy Pylance false positive in wakeword_bench.py 2026-06-09 16:02:50 +02:00
wakeword.py feat: dedicated wakeword models for cancel and send actions (v2.03.30) 2026-06-10 12:07:03 +02:00