fix(voice): CLI/TUI voice mode UX and environment fixes (15 items) - #73520

Merged

teknium1 merged 23 commits into

mainfrom

fix/voice-cli-tui-ux

Jul 28, 2026

Commits on Jul 28, 2026

chore: map salvage contributors (voice CLI/TUI UX batch)
teknium1
committed
d49e572
fix(cli): disarm continuous voice mode via hotkey during agent/transcribe (#67545)

kyssta-exe 25470058+kyssta-exe@users.noreply.github.com
authored and
teknium1
committed
e45ab7e
fix(voice): prevent restart race when continuous mode stops after 3 no-speech cycles

brunopirz
authored and
teknium1
committed
58e5aab
fix(voice): enforce voice.max_recording_seconds (was dead config)

Solitud1nem
authored and
teknium1
committed
cae71ba
fix(voice): thread max_recording_seconds through the TUI path, add behavior coverage

Solitud1nem
authored and
teknium1
committed
30fb223
fix: play returned TTS audio path in CLI voice mode

valda
authored and
teknium1
committed
21d9a1b
fix(voice): prefer requested MP3 for playback

valda
authored and
teknium1
committed
68ad4e5
feat(voice): make beep notification volume configurable from config.yaml

Kewe63
authored and
teknium1
committed
d93fadc
fix(voice): capture at the input device's native sample rate

bricelb
authored and
teknium1
committed
024febe
fix(voice): honor forwarded audio (PIPEWIRE_REMOTE) in WSL detection

Solitud1nem
authored and
teknium1
committed
9c8da70
fix(voice): add WSL audio warmup to eliminate RDP crackling

kharitonov-ivan
authored and
teknium1
committed
60904df
fix(voice): add WSL2 PowerShell audio fallback for TTS playback

ygd58
authored and
teknium1
committed
40764e9
fix(voice-mode): make Termux:API detection robust against pm probe failures

xxxigm
authored and
teknium1
committed
38a992f
test(voice-mode): pin Termux:API detection probe ladder + #31015 fallback

xxxigm
authored and
teknium1
committed
f0df21f
fix(voice): skip sounddevice on macOS to avoid TCC media-library prompt

authored and
teknium1
committed
6920715
voice: one macOS output policy across WAV, beeps, streaming TTS + tests

authored and
teknium1
committed
15c91bc
fix(voice): point pip install hint at venv pip instead of system Python

RedClaus
authored and
teknium1
committed
d1d887f
fix(docs): target installer venv in voice extra install command

liuhao1024
authored and
teknium1
committed
4cd9ada
fix(cli): surface local STT model preparation

locker95
authored and
teknium1
committed
dff2d4d
fix(cli): only prefix voice-transcribed messages with the voice-input instruction (#65827)

webtecnica
authored and
teknium1
committed
9077173
fix(voice): honor quoted beep_enabled strings + correct PortAudio OSError hint
teknium1
committed
ba261a2
test: adapt salvaged voice tests to current main + lint fix
teknium1
committed
bdb600b
test: update voice-submission test for _VoiceInputMessage sentinel (#65827)
teknium1
committed
ff13dd3

Navigation Menu

Uh oh!

fix(voice): CLI/TUI voice mode UX and environment fixes (15 items) - #73520

fix(voice): CLI/TUI voice mode UX and environment fixes (15 items)#73520
teknium1 merged 23 commits into
mainfrom
fix/voice-cli-tui-ux

Commits on Jul 28, 2026

chore: map salvage contributors (voice CLI/TUI UX batch)

fix(cli): disarm continuous voice mode via hotkey during agent/transcribe (#67545)

fix(voice): prevent restart race when continuous mode stops after 3 no-speech cycles

fix(voice): enforce voice.max_recording_seconds (was dead config)

fix(voice): thread max_recording_seconds through the TUI path, add behavior coverage

fix: play returned TTS audio path in CLI voice mode

fix(voice): prefer requested MP3 for playback

feat(voice): make beep notification volume configurable from config.yaml

fix(voice): capture at the input device's native sample rate

fix(voice): honor forwarded audio (PIPEWIRE_REMOTE) in WSL detection

fix(voice): add WSL audio warmup to eliminate RDP crackling

fix(voice): add WSL2 PowerShell audio fallback for TTS playback

fix(voice-mode): make Termux:API detection robust against pm probe failures

test(voice-mode): pin Termux:API detection probe ladder + #31015 fallback

fix(voice): skip sounddevice on macOS to avoid TCC media-library prompt

voice: one macOS output policy across WAV, beeps, streaming TTS + tests

fix(voice): point pip install hint at venv pip instead of system Python

fix(docs): target installer venv in voice extra install command

fix(cli): surface local STT model preparation

fix(cli): only prefix voice-transcribed messages with the voice-input instruction (#65827)

fix(voice): honor quoted beep_enabled strings + correct PortAudio OSError hint

test: adapt salvaged voice tests to current main + lint fix

test: update voice-submission test for _VoiceInputMessage sentinel (#65827)