Short tones for joining and leaving a voice room, other participants
joining and leaving, new messages outside the viewed chat, mentions,
microphone mute and unmute, deafen and undeafen, server connection and
loss, and errors. The tones are generated at runtime (rising motifs for
arrival or enable, falling for leave or disable) with an attack and a
fade-out so they never click; no third-party audio is bundled.
Inside a voice room the sound is mixed into the open playback stream,
otherwise a short one-shot stream is opened on the selected output device
off the UI thread, capped at two at a time. While deafened nothing plays
except the undeafen sound. Settings: master switch, volume and separate
switches for voice room, messages and mentions, stored with the audio
preferences, with a new settings section and a test button.
Switching from room A to room B replaced the voice state with a joining
state while the microphone was still capturing in A, and a deleted room
left the capture pipeline running with no visible state.
A pending join is now stored separately, so the live session stays shown as
connected until the server acknowledges the new room or refuses it. The
capture pipeline is stopped whenever it no longer belongs to the current
voice target, whatever a pending join is waiting for.
The status bar and voice page kept saying the room was being joined while
the sidebar already reported a connected voice. Clicking the room you were
already in set a free-text 'joining' message that nothing cleared, because
the server does not answer a repeated join, and the voice page showed its
joining label whenever the member list was empty.
Joining is now an explicit voice state. The sidebar, the voice page and the
status text all read one derived status; a repeated click only shows the
room, and a server error cancels a pending join.
Capture and playback failures were only written to the log on detached
threads, so a microphone or output that could not be opened still showed a
connected voice room with no sound. Threads now report into a shared
diagnostics sink that the UI polls: errors appear in the status bar, a
room with neither working half is marked failed, and signal counters
(captured, sent, received, decoded, input peak) are logged every ten
seconds. A microphone that delivers only digital silence is reported, and
rejected voice datagrams are logged instead of dropped silently.
Voice activation now tracks the background noise floor and opens just
above it, capped by a much lower default ceiling; the old fixed threshold
(about -26 dBFS) rarely opened for normal speech. Stream configuration only
accepts f32 formats and says so when a device offers nothing else.
Playback gains a mixing queue for short notification effects.
The device pickers showed every ALSA PCM plugin (Discard all samples, Rate
Converter, JACK, Open Sound System, surround/dmix aliases). Enumeration now
goes through a pure filter that keeps the system default, the PipeWire and
PulseAudio servers and one entry per hardware endpoint, with technical tails
removed and a stable sort.
Selections are stored by device id instead of list position, so the filter
and hotplug cannot move a saved choice. Indexes saved by older versions are
migrated by device name; a missing device falls back to the default.
The same preferences change records whether the voice mode was chosen by
the person: a profile that only carries the old push-to-talk default now
starts in voice activation, explicit choices are kept.
Replace the custom X25519/HKDF/ChaCha20 TCP layer with rustls (TLS 1.3 only,
ALPN vnox-lnex/3, no plaintext or fallback). Add a trust store (pin, trust on
first use, public CA), block changed fingerprints and identities, bind the
Ed25519 auth signature to the TLS exporter, add RU/EN trust dialog and
update the interop test and e2e voice tool.