🤖 This fixes two problems with the same cause. Since #3268, the relay
can answer an unpinned `/query` from the Aurora read replica, as long as
the replica has proved it's within `BUZZ_REPLICA_READ_MAX_AGE_MS` (1000
ms) of the writer. That's fine for display reads, but it doesn't
guarantee a client sees its own write.
**Problem 1: channel creation (Wes's report).** Right after creating a
channel, the desktop read back the channel's kind:39000 metadata by
`#d`. If that read landed on a replica that hadn't replayed the event
yet, it failed with `channel created but metadata not yet available`.
The channel existed, but the chosen template (canvas, agents) was never
applied.
**Problem 2: member lists.** After any membership change, the
member-list refetch could also hit a lagging replica. The relay returned
the list from before the change, and the desktop cached it as fresh for
five minutes (`CHANNEL_MEMBERS_STALE_TIME_MS`). Someone just removed
could still show as a member, or someone just added could be missing.
## Fix 1: metadata read-backs go to the writer
`channel_metadata_filter` is a shared builder that adds `"consistency":
"strong"`, so the relay serves the read from the writer. It follows the
same pattern as the canvas head reads. All four metadata read-backs that
follow a write use it: `create_channel`, `update_channel`,
`fetch_starter_channel_metadata` (its retry and fallback are unchanged)
and the open-DM refetch in `dms.rs`. A unit test fails if the flag is
removed from the builder.
The template's agent-add step also reads channel members (kind:39002)
right after create. `get_channel_members` takes an opt-in
`read_your_writes` flag that sends `"consistency": "strong"`. Only
`createChannelManagedAgents` sets it on that path, so the agent-add step
sees the creator's membership.
## Fix 2: a 31-second writer-read window after membership changes
Every membership write the desktop makes calls
`noteChannelMembershipChange` for each channel it changed:
- the native write wrappers `createChannel`, `openDm`, `joinChannel`,
`leaveChannel`, `addChannelMembers`, `removeChannelMember` and
`changeChannelMemberRole`
- the Bestie conversation open
- huddle agent add and remove, for both the huddle channel and its
parent
- `ensure_starter_channels` and `sync_agents_to_active_huddle`. Both
return every channel they created, joined or changed in
`changed_channel_ids`, including ones that landed before a later step
failed. The original error text is rethrown unchanged.
- live join, leave, removed and `admin_kick` system messages, and
member-added and member-removed notifications
- a report kick resolved from the Admin Console, once the relay confirms
it succeeded and only when the report's channel is in the active
community. The kick goes through the admin HTTP API, not a channel
event. Ban, timeout and their lifts only change `community_bans`, not
channel membership, so they need no record.
For 31 seconds after a change, measured on `performance.now()`,
`getChannelMembers` for that channel sends `readYourWrites`. Why 31
seconds: the relay caps its replica staleness budget at
`FENCE_STALENESS` (30 s), so once the window closes, no supported
configuration can serve the pre-change list. The constant's comment ties
it to `FENCE_STALENESS`. One app-level listener cancels any roster fetch
already in flight and refetches. When it registers, it also refreshes
every channel whose window is still open. All other member-list reads
still use the replica.
**Community switch.** The window map is cleared when the community
changes. Each awaited write captures a generation through
`beginChannelMembershipWrite()` before its first native call, and the
reset advances it. So a write that settles after a community switch
records nothing in the new community. Live events don't need this fence,
because the reset disconnects the relay client first and frames from the
old connection are dropped.
## Coverage
- A Rust test drives `ensure_starter_channels` against a stub relay that
accepts the creates but never serves metadata, and checks that both
created starter channels are still reported.
- Desktop tests cover the writer-read window, the Admin Console kick,
and writes that settle after a community switch.
- The desktop E2E test bridge (`desktop/src/testing/e2eBridge.ts`)
mirrors the strong reads and `changed_channel_ids`, so E2E runs exercise
the fixed behavior.
This doesn't address the smaller race in the relay's per-pod
accessible-channels cache, and it doesn't change mobile.
---------
Signed-off-by: Will Pfleger <pfleger.will@gmail.com>
Co-authored-by: Duncan <dcfd242e557282d7a1e2cf2e6877522682f1e5c6156dc92ca7d90eaedd3b0f95@buzz.block.builderlab.xyz>
🤖 I’m Larry, updating this description on Logan’s behalf.
## Summary
Build named macOS demo apps without Finder automation or collisions with
installed Buzz. `just desktop-demo-build "PR 6407 Demo"` produces a
matching app and DMG, with a fresh build identity even when the same
display name is reused.
- The headless DMG packager uses `hdiutil`; optional Finder styling is
bounded. The existing production release recipe is unchanged.
- Each demo has independent app data, keychain, nest, CLI name,
voice-model storage, repository discovery, and agent OAuth/config
storage. Reset preserves production and sibling-demo state, and retains
retry intent when credential removal or root resolution fails.
- Native links accept only the active build’s registered scheme, then
translate validated entity links into the frontend’s canonical `buzz:`
format.
- The recipe builds all six executable sidecars. Display names are
capped at 31 ASCII characters so the generated identity fits Rust’s
build-time limit.
**Open delivery requirement:** downloaded demos must run without a
Gatekeeper security override. The current recipe is ad-hoc signed and
unnotarized; it does **not** satisfy this requirement. Trusted
branch-demo signing/distribution remains blocked on establishing an
approved signing path. This PR is not being presented as complete
download-and-run delivery.
### Related issue
N/A — reported in the Buzz DMG-packaging workstream.
### Testing
At `11ce21ff97cb387ad676e7caa65b00964097d0bb`, macOS Blox passed the
Tauri workspace suite and compiled-flags gate (including the full
named-demo state; each library pass: 2,992 passed, 19 ignored), Tauri
all-target clippy, the full `buzz-agent` package suite, and frontend
lint/typecheck plus 5,733 tests. Regression coverage includes
cold-start/running entity-link handling, wrong-build rejection, OAuth
deletion failure and retry, unresolved credential roots, and
production/sibling preservation.
At the same head, an extra full named-demo/mesh-enabled run had 3,092
passing tests and one failure: a pre-existing shared-compute `auto`
versus `mesh` expectation, also reproduced on the old published head
`a77b25eca`. The ordinary and demo-state matrix above passes; this is
not an all-features-green claim. Live macOS Launch Services delivery
remains unverified.
GitHub CI completed with 30 successful checks and 9 skipped. The
exact-range security review has not run; its authorization notice
remains open. CI success does not establish trusted signing or
downloaded-app launch.
Earlier demo artifacts established matching app/DMG names, side-by-side
launch, and six non-empty executable arm64 sidecars. These screenshots
show an earlier artifact, not a new build of the final repair commit.
Signature-integrity checks are not Gatekeeper/notarization evidence.
<img width="1032" height="548" alt="Buzz PR 6407 Demo disk image
containing the matching app"
src="https://github.com/user-attachments/assets/bca0277e-db03-4308-b280-fcad55e6d601"
/>
<img width="1186" height="821" alt="Buzz PR 6407 Demo running alongside
other Buzz installations"
src="https://github.com/user-attachments/assets/b4bf4ae5-c341-4e15-8090-9d2ea7c623b6"
/>
---------
Signed-off-by: Logan Johnson <loganj@squareup.com>
Signed-off-by: Larry <8cf5a83f590ec0955b11647d1c88f796a98e088c30a492c58e0e46c3026ae7a4@buzz.block.builderlab.xyz>
Co-authored-by: Other Brother Darryl <cee32d92756729ee0c097c5661b879c6199931cd25315c8cf398dcbf0f155cf1@buzz.block.builderlab.xyz>
Co-authored-by: Larry <loganj+sandbox-larry@squareup.com>
Co-authored-by: Larry <8cf5a83f590ec0955b11647d1c88f796a98e088c30a492c58e0e46c3026ae7a4@buzz.block.builderlab.xyz>
## Summary
- ignore mobile room tone when deciding whether human speech should
interrupt agent audio
- show a subtle waiting animation while an agent prepares a response,
then restore its avatar when speech begins
- render agents from Huddle membership before their audio peer starts
transmitting
## Testing
- `just ci`
- signed Release build installed and launched on a physical iPhone
- live Huddle voice and agent-membership behavior verified on device
---------
Signed-off-by: kenny lopez <klopez4212@gmail.com>
Co-authored-by: Mongo <9cfd347903944d5b85aa6c93d2ab67381b978a92a31914bca69998968752a1d7@buzz.block.builderlab.xyz>
## Problem
Humans cannot interrupt agents in huddles: agent TTS keeps playing over
a talking human (reported by @tlongwell-block, 2026-08-20). The full
requirement: **any human talking — local or remote, any input mode, any
audio rig — must interrupt any agent on the huddle.**
## This is a restoration, not a new feature
- **`b29c8cdaa` (#4281, 2026-08-04) deleted working local VAD
barge-in.** The pre-image shipped `BARGE_IN_DEBOUNCE_FRAMES = 20` (320
ms sustained speech cancels TTS), live at the production call site
(`pipeline.rs` passed `Some(tts_cancel)` unconditionally). The same day,
`ce3cf3cd2` (#4694) flipped the default input mode from VAD to
push-to-talk, which masked the loss.
- **`068a83b09` (#5671, 2026-08-13) removed the remaining TTS-awareness
plumbing from STT** (deliberately, to keep transcribing over agent audio
— a good change). Consequence: this PR is a re-plumb through the
`PlaybackCoordinator` from #6341, not a revert.
- A gap that **never** worked is also closed: PTT-mode users who open
the mic via the mute button (`manual_mic_unmuted` postdates the
deletion) transcribed fine but could not barge in.
## Design
All floor state lives under the single `PlaybackCoordinator` lock —
onset acceptance, epoch bump, synthesis invalidation, player
replacement, and output lease are one committed transition. No cancel
flag can be observed out of order with the playback state it describes.
- **Human floor**: local + per-peer remote ownership with epoch
invalidation. Onset cancels playback by queue replacement; late
synthesis for a stale epoch cannot append or restart.
- **Local onset (VAD)**: on an **isolated output route** (all CoreAudio
output-stream terminals report headphones), a confirmed short onset
interrupts immediately — no echo path exists. On a **coupled route**
(speakers, unknown, virtual, mixed), the restored **20-frame / 320 ms
sustained-speech debounce** discriminates a real human from speaker
bleed; the deleted code's comment records that 80 ms was tried and
false-triggered on laptop speakers. Route classification is queried
fresh at each onset (never cached — default-device re-routing mid-huddle
would strand a stale verdict).
- **Mic-open gate is per-frame**: barge-in observes on any frame where
the mic is actually open — pure VAD mode, or PTT with the mic manually
unmuted (key-held frames defer to the shortcut's own cancel).
- **Remote onset**: sustained non-DTX frames from a peer enter the same
persistent floor (independent of whether playback is live — a human
speaking while TTS is idle blocks late-arriving synthesis from starting
over them). Release on sustained DTX/absence, peer departure, and
recv-loop exit, with guards so a vanished peer cannot wedge the floor.
- **Output lease**: accepted appends renew an `Active` lease;
drain/cancel/onset start a 100 ms tail hangover (conservative against
measured ~12 ms/~1 ms CoreAudio tails), so speaker-tail bleed in the
just-drained window cannot self-trigger the coupled path.
## Known limitations (phase 2 pointers)
- Coupled-route mid-output barge-in pays the 320 ms debounce; a
playback-reference echo discriminator would shorten it.
- A speakers-rig participant's bleed can enter their mic and hold the
floor for other machines (bounded by release debounce).
- Non-macOS routes classify as coupled (fail-safe).
## Verification
- Full `buzz-desktop --lib` suite at head `b0459ae4a`: **2707 passed, 0
failed, 18 ignored** (pinned cargo 1.95.0).
- Exact CI recipe `just desktop-tauri-clippy`: PASS at head; base arm at
merge-base `b728a2af3` confirms the two
`#[allow(clippy::too_many_arguments)]`s cover branch-caused threshold
crossings (human_floor threading), not inherited noise.
- Regression tests pin: 20-frame threshold + reset-on-gap, short-coupled
rejection, sustained-coupled acceptance, coupled-idle acceptance,
remote-idle delayed-TTS rejection, output-tail hangover boundary (during
= rejected, after = accepted), isolated onset, per-frame mic-open gate
truth table, PTT+manual-unmute sustained coupled acquisition.
- `LocalBargeIn::observe` is covered as two joined halves (gate truth
table in `local_barge_in.rs`, floor transition in `tts_playback.rs`);
its body is a straight-line wrapper around a live CoreAudio query, left
uninjected deliberately.
- Coverage precision (mutation-verified):
`manual_open_ptt_sustained_speech_acquires_coupled_floor` pins the gate
→ 20-frame debounce → acquire → floor-blocked chain on the
coupled-**idle** cell. The live-output override leg is carried by
`sustained_coupled_speech_overrides_live_output_suppression`
(tts_playback.rs); the joiner's `sustained_coupled` argument is not
load-bearing there (flipping it to `false` leaves the test green, while
shortening the debounce by one frame turns it red).
- Live arms in progress: pre-regression build `b29c8cdaa^` staged to
confirm the deleted mechanism worked; two-endpoint remote-leg test
pending a second human.
## Commits
1. `fb681a5a9` — restore human barge-in (coordinator floor, lease, route
isolation, remote floor, 320 ms coupled debounce)
2. `b4265418e` — enable barge-in for manually opened mics (per-frame
gate; closes the PTT-open-mic gap)
3. `886489f2b` — extract local barge-in policy module (file-size
ratchet; also hoists the CoreAudio route query from per-frame to
per-onset, named in the commit message)
4. `b0459ae4a` — two targeted clippy allows for the widened worker
signatures
## Credits
Built by **Wren**. Regression archaeology and the PTT-open-mic gap by
**Dawn** (who also killed her own first fix as vacuous and caught a
clippy blocker before it hit CI). Review blockers (coordinator
serialization, idle-onset floors, output lease) by **Mari**. Live rig
verification by **Max**. Coordination and verification by **Eva**.
Opened by Eva with Tyler's explicit direction; commits carry agent
trailers.
---------
Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
## Problem and intent
Buzz huddles have been clipping speech into fragments: one live turn
arrived as the single letter “M,” and another ended mid-sentence. The
detector was making each frame decision independently, but the
surrounding endpointing policy had no onset confirmation or pre-roll,
used one threshold for both entering and leaving speech, and silently
discarded short segments.
This PR keeps Earshot 1.1.0 and fixes that policy around it. It is
intentionally model-independent so the separate Earshot/Silero bake-off
can evaluate detectors under the same segmentation behavior.
## What changed
One production file changed: `desktop/src-tauri/src/huddle/stt.rs`
(**+367/−79** versus `main`).
- Add a pure `VadEndpoint` state machine around Earshot probabilities.
- Preserve 256 ms (16 frames) of pre-roll before confirmed onset.
- Require three consecutive frames above 0.50 to enter speech.
- Use 0.35 to leave speech, preserving hysteresis-band audio.
- Retain 96 ms (6 frames) of hangover while keeping the existing 304 ms
silence-flush window.
- Make short-segment drops visible in logs instead of silent.
- Add boundary, onset, hysteresis, hangover, drop-path, and PTT policy
tests.
- Bind the thresholds to Earshot 1.1.0 in source. Earshot 1.2.x is
deliberately parked pending a matched-policy bake-off; #6392 separately
prevents Renovate from silently crossing that boundary.
## Boundary behavior and known tradeoffs
A hard message boundary clears pre-roll. That prevents segment N audio
from reaching segment N+1, but a fast follow-up turn may receive less
than the full 256 ms onset window.
On the 121-clip corpus, **12 of 25 non-first segments had truncated
pre-roll, with 32 ms worst observed at the shipping constants**. This
result **holds at both 208 ms and 256 ms on this corpus**; it is not
independent of pre-roll. The observed follow-ups were bimodal—fast cases
clustered at 2–10 frames and the next case was 19—so no larger pre-roll
budget could reach those fast cases in this corpus.
If a future refactor makes pre-roll survive a boundary, segment N
reaches segment N+1 under the strict predicate:
```text
gap < VAD_PRE_ROLL_FRAMES - VAD_ONSET_FRAMES
```
At the shipped values that means gaps 0–12 leak and gap 13 is the first
clean case. Hangover and the silence-flush window do not enter this
bound because leakage rides the pre-roll deque. Marker-origin tests
exercise the real flush/reset and next-onset drain; a rolling-buffer
mutant leaks 3,328 samples and fails them.
Other known behavior:
- `voiced_frames` now counts frames above the 0.35 exit threshold once
speech begins, not only frames above 0.50. Corpus drop counts measured
under `> 0.50` counting must be re-baselined rather than compared
directly; the old “19 silent drops” figure is not a valid before/after
baseline.
- If PTT releases while the manually-open microphone flips or remains
on, there is no combined-transmit falling edge. The uninterrupted
utterance correctly closes on normal VAD timing instead of an edge
flush.
- The GUI-to-`push_audio_pcm` wiring is unchanged. The final acoustic
gate exercises the checked-in release-profile latency harness against
the real production `SttPipeline` and `TtsPipeline`; Tyler’s hand-test
covers the live GUI leg by design.
## Verification ledger
All gates attest exact head `2dad6bb0e2e2a0af115a9a06c26cf6183fedb633`.
- **Implementation:** complete pre-push gate green; remote SHA matched
local HEAD.
- **Exact-SHA verification:** `cargo fmt --check`; `cargo clippy
--all-targets -- -D warnings`; full `desktop/src-tauri` tests: **2,702
passed, 0 failed, 18 ignored**; HEAD re-confirmed unchanged afterward.
- **Mutation review:** engine region SHA-256
`3c161d8f6e4d22c02dff3c76498430964beda51a034d34651dbdb03e8ff69aa0`; 5/5
policy mutants killed, including deleted-clear and rolling-buffer
regressions. Direct gap sweep leaked at 0–12 and was clean from 13.
- **Release acoustic gate:** negative control deliberately overstated
the expected segment count and failed with exit 101. Soft/short onset
produced 1/1 segment (`“I'm happy.”`). Natural-pause fixture used two
exact 700 ms pauses (>304 ms) and produced the pre-registered 3/3 intact
segments at the scripted boundaries.
- **Release CPU beside Pocket TTS:** soft fixture median 32.1%, p95/max
56.8% (5 sparse samples); natural fixture median 14.55%, p95/max 17.2%
(8 sparse samples). These are scoped process samples from synthesis
start through append acceptance, not a general desktop CPU benchmark.
Local review receipts (not committed to the repository):
- `.scratch/vad-live-release-2dad6bb0e/`
- `RESEARCH/VAD_ARM_B_MUTATION_RIG_2026_08_20/`
- `RESEARCH/EARSHOT_1_1_0_TO_1_2_2_MEASUREMENT_2026_08_20.md`
## Hand-test focus
Before merge, exercise the two live shapes that originally failed:
1. Soft short openers such as “M” and “yes.”
2. Natural mid-sentence pauses.
Also try a reply immediately after the previous message commits; that is
intentionally the least-protected onset case because hard boundaries
clear pre-roll.
---
Authorship disclosure: this change was implemented and the PR opened by
**Wren**, Tyler’s Buzz agent, using Tyler’s GitHub identity after prior
disclosure and authorization in the originating Buzz thread. Tyler
remains the accountable human reviewer/operator.
---------
Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
## Summary
- introduce a `PlaybackCoordinator` as the exclusive owner of the rodio
`Player`
- serialize append, cancellation, synthesis-flight, and drain-release
state
- cancel without blocking the coordinator by replacing the player queue
- preserve first-append/onset bookkeeping across cancellation
- increase the existing playback lead-in from 20 ms (480 samples) to a
derived 22 ms (528 samples), covering rodio 0.22.2's 512-source-sample
add-time bootstrap span
This is the behavior-preserving coordinator prerequisite for the
follow-up floor-authorization policy. The only intentional playout
difference is +2 ms silence at playback-chunk starts to prevent the
measured fresh-player bootstrap chirp.
## Verification (all at this exact head,
`8bfa4c0531834dce39f5ad0266f2674505e659bf`)
- `cargo clippy --manifest-path desktop/src-tauri/Cargo.toml --workspace
--all-targets -- -D warnings` — clean
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml --package
buzz-desktop` — 2,596 passed / 17 ignored / 0 failed
- coordinator concurrency tests: stalled cancellation, append-vs-cancel
serialization, cancel election, first-append rearming, and the post-swap
`starts_playback_chunk` invariant
- headless known-answer sweep: 528-sample lead-in clean across 7/7
device configurations (incl. falsifier arm at 192k/96k/32k/16k/8k and
1/2/4/6 ch); current 480 dirty across 5/7
- live CoreAudio gate at production buffer shape (24 kHz mono, real
`build_sentence_append_buffer` path): bootstrap chirp eliminated (660.1
Hz head vs 660 Hz control), no false drain, cancel-to-silence ~1 ms vs
12 ms baseline
- independent line review + mutation testing on the shipping coordinator
bytes: 4/6 mutants killed, 2 shown equivalent under rodio 0.22.2
`Player::drop` semantics
## Playout cost
- +2 ms lead-in at each playback-chunk start versus main
- first post-cancel utterance receives 22 ms total intentional silence
(not stacked cushions)
## Provenance
Authored and reviewed by Buzz agents (Wren: implementation; Dawn: line
review, known-answer control, mutation testing; Max: live CoreAudio
gates; Eva: coordination and final verification), with review records on
the Buzz relay. This PR was briefly merged to main directly by mistake
and reverted via ref reset; this is the same commit going through the
front door.
---------
Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Signed-off-by: Dawn <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Dawn <c6237ef84fa537c78dcee78efd2d4e59f728859c7f194da42ac51ededfa0be05@buzz.block.builderlab.xyz>
## Overview
**Category:** feat (env-gated experiment + one exact always-on
optimization)
**Problem:** Speech-end -> first TTS audio through the desktop huddle
pipeline measures **924–1087 ms** on an M4 Max with a 0 ms LLM leg.
Voice turns feel sluggish no matter how fast the agent replies. Baseline
breakdown: ~300 ms hardcoded VAD silence flush + ~150–250 ms Parakeet
decode + ~380–550 ms TTS synthesis before the first player append.
**Outcome:** With all levers enabled, e2e time-to-first-audio measures
**347–384 ms** (307–357 ms on a longer utterance) on the same hardware,
harness, and production pipelines. Defaults preserve production behavior
everywhere except one deterministic, bit-exact cache win.
## What's in here
### Levers (all default-off, env-gated)
| Lever | Env | Effect (measured medians, short utterance) |
|---|---|---|
| Speculative Parakeet decode | `BUZZ_STT_SPECULATIVE=1` | STT leg ->
~max(flush, decode) |
| Streaming TTS synthesis | `BUZZ_TTS_STREAMING=1`,
`BUZZ_TTS_EMIT_FRAMES` | first audio 380–550 -> 211–320 ms (emit=12,
bit-exact) |
| ONNX intra-op threads | `BUZZ_STT_THREADS`, `BUZZ_TTS_THREADS` | TTS
first audio 211–320 -> 129–180 ms (4 threads) |
- **Speculative decode** starts the Parakeet decode at the *first*
silent VAD frame, overlapping it with the flush window. Resumed speech
invalidates the result (voiced-frame-count check); held silence emits it
instantly at the flush boundary.
- **Streaming TTS**: new `synth_chunk_streaming` (buzz-voice)
interleaves the Flow LM frame loop with incremental *stateful* Mimi
decoding, emitting PCM deltas to the player via the existing
`PlaybackChunkAudio` decoration. At `emit_frames=12` (the decoder's
native chunk) streamed audio is **bit-identical** to the batch path —
verified by the ignored test
`incremental_stateful_decode_matches_batch_decode` (max|diff|=0).
Smaller deltas are faster but diverge (~23 dB SNR; decoder intra-chunk
lookahead), hence the default of 12.
> **Removed after live testing:** the `BUZZ_STT_FLUSH_MS` flush-window
override. Lowering the silence window below natural mid-sentence pauses
(the fast-path recipe said 150 ms) split single spoken sentences into
multiple messages and confused the listening agents. The window is a
turn-taking quality knob, not a latency lever — it is now fixed at the
production 300 ms value.
### Push-to-talk grouping fix (always-on)
A held push-to-talk shortcut is an explicit "I am not done talking"
signal, so silence never ends the utterance while it is held — even when
the microphone is also manually open. The utterance flushes on shortcut
release (existing transmit-edge flush); a manually open mic with the
shortcut up keeps normal VAD pause flushing. Gate is the pure
`vad_flush_allowed` function with a unit-test truth table.
### Always-on (exact): voice-conditioning cache
Phase profiling (`BUZZ_TTS_PHASE_LOG=1`) showed a fixed ~160 ms
`condition_voice` Flow-LM pass on *every* chunk, re-deriving the same
post-conditioning state for the same reference voice. The state is now
snapshotted after first computation and restored per chunk (dtype-tagged
tensor copies, keyed identically to the existing `cached_voice`).
Deterministic — same tensors in, same tensors out. The default path's
TTS leg drops from 380–550 ms to 225–355 ms with no configuration.
### Bench harness
`huddle::latency_bench` (`#[cfg(test)]` + `#[ignore]`) drives the real
`SttPipeline` and `TtsPipeline`, feeding a 48 kHz WAV in real-time 100
ms batches (AudioWorklet cadence) with a configurable fake LLM in place
of the relay leg, timing speech-end -> transcript -> speak() -> first
accepted player append.
```
BUZZ_STT_SPECULATIVE=1 BUZZ_TTS_STREAMING=1 \
BUZZ_TTS_THREADS=4 BUZZ_STT_THREADS=2 \
BUZZ_BENCH_WAV=<48k f32 mono wav> \
cargo test --release -p buzz-desktop --lib huddle::latency_bench -- --ignored --nocapture
```
## Tradeoffs to weigh before promoting any lever to a default
- **Speculative decode**: the speculative buffer has ~1 silent tail
frame vs ~19; observed one CTC wobble ("fail" vs "failed") in 24 turns.
Mitigation if productionized: zero-pad the speculative buffer to match
the flush-path shape.
- **Threads**: defaults stay 1 pending the min-spec (4-core Intel) A/B
flagged in the existing `STT_NUM_THREADS` comment.
- **Streaming at emit<12** is NOT the same waveform — don't ship below
12 without an ear pass.
## Validation
- Full desktop lib suite: **2408 passed / 0 failed** at this head
(`18fab2e1c`).
- buzz-voice suite green; bit-exactness test passes against the
production batch decode.
- Defaults-only bench rerun stays in the baseline family everywhere
except the exact conditioning-cache win (stt 525–532, tts 225–355).
- `cargo clippy --workspace --all-targets -- -D warnings` + fmt clean
(pre-push hook battery green).
Measurement notes with per-lever logs: Eva's workspace,
`RESEARCH/HUDDLE_E2E_LATENCY_OPTIMIZATION_2026_08_12.md` +
`RESEARCH/HUDDLE_E2E_STT_FAKELLM_TTS_BASELINE_2026_08_12.md`.
## Suggested promotion order
1. Conditioning cache (in this PR, always-on, exact).
2. Streaming TTS at emit=12: bit-exact audio, biggest UX win — needs the
env-gate removed + barge-in soak + an ear pass on a real huddle.
3. Speculative decode with silence padding: near-free ~100–150 ms.
4. Threads: after min-spec A/B.
---------
Signed-off-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Signed-off-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Signed-off-by: Max <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Signed-off-by: Mari <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Signed-off-by: tlongwell-block <tlongwell@block.xyz>
Co-authored-by: Eva <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: Wren <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: Max <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Co-authored-by: Mari <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Co-authored-by: tlongwell-block <tlongwell@block.xyz>
## Overview
**Category:** improvement
**User impact:** Link previews appear in the composer and travel as
privacy-safe sender-authored snapshots, so recipients never contact the
linked site merely by opening a conversation.
**Problem:** Cold-cache link paste could freeze the composer before the
URL painted; recipient-side unfurling leaked visits; invalid or
unresolved preview work could interfere with sending or leave dead cards
behind.
**Solution:** Paint pasted links before starting cold resolver work,
resolve only in the sender's composer, attach only complete validated
snapshots at Send, and render authored snapshots without recipient
fallback fetching.
## Behavior
- **Cold paste stays responsive:** bare and angle-bracket URL paste
paths commit the visible link before resolver work begins.
- **Sender-only fetching:** metadata is resolved while composing;
recipients render only the sender-authored snapshot.
- **Send never waits:** pending, failed, invalid, and unsendable
previews are omitted. They do not block or cancel the message.
- **Terminal misses disappear:** failed, timed-out, or 404 resolver
results remove the composer card while preserving visible link text.
- **Display-text links work:** Markdown links such as `[review the pull
request](…)` produce and send the same snapshots as bare URLs.
- **Compact and Rich presentation:** Compact remains the default; Rich
preserves source description line breaks and paragraphs.
- **Immediate draft-wide dismissal:** clicking × immediately hides all
previews for the draft, suppresses links pasted later, and emits only
`["link-preview", "none"]`. No confirmation detour. Suppression resets
after send or clearing the draft.
- **Zero recipient fallback:** missing, stale, malformed, off-relay,
unsupported, or suppressed snapshots remain ordinary visible links;
recipients never regenerate them.
## Implementation
- Resolve previews from deferred composer URL state so paste can paint
first.
- Upload finished preview media to the active community relay and
snapshot only valid, sendable media references.
- Atomically capture ready snapshots at submit time; never append a late
preview after send.
- Validate snapshot and suppression tags in desktop/native and relay
ingestion, rejecting duplicate or mixed forms.
- Render composer previews as stable 55px attachment cards at desktop
and narrow widths.
- Add deterministic E2E coverage for cold paste,
ready/pending/failed/invalid previews, display-text links, multiline
Rich descriptions, immediate dismissal, later-pasted links, and
suppression reset.
## Validation
Validated head: `9807ba8952f190e76153834abf8ab61dd40be5e2`
- Push hooks passed: `check-push-org`, branch skew, desktop check,
mobile tests, desktop tests, Rust tests, and desktop Tauri checks.
- Focused screenshot E2E at the validated head: 5/5 passed across
Compact/Rich composer and recipient states, 800px/420px geometry,
display-text links, multiline descriptions, and immediate dismissal.
- PR CI was triggered for this exact head and is currently running;
completed checks are green at the time of this update.
- Worktree is clean and both PR head and validated branch resolve to
`9807ba895…`.
## Screenshots
### Compact composer
| Loading | Ready |
|---|---|
| 
| 
|
### Rich composer
| Loading | Ready |
|---|---|
| 
| 
|
### Responsive composer
| 800px loading | 800px ready |
|---|---|
| 
| 
|
| 420px loading | 420px ready |
|---|---|
| 
| 
|
### Recipient presentation
| Compact | Rich |
|---|---|
| 
| 
|
### Display-text Markdown link
| Composer | Recipient |
|---|---|
| 
| 
|
### Rich multiline description
### Immediate dismissal
| Before × | Immediately after × |
|---|---|
| 
| 
|
---------
Signed-off-by: Taylor Ho <taylorkmho@gmail.com>
## Summary
- add a visible Stop control for interrupting agent speech
- make push-to-talk available by default while preserving manual mute
controls
- refine agent management, muted audio states, drawer layering, and
return navigation
- suppress duplicate notification sounds for Huddle messages
## Why
Huddles could trap users behind long agent speech, hide useful agent
controls, and leave temporary Huddle state visible after the call. The
drawer also regressed when the terminal substrate began painting behind
the rounded app surface.
## Validation
- `just desktop-ci`
- focused Huddle Playwright coverage for the drawer, speech
interruption, agent picker, and leave navigation
---------
Signed-off-by: kenny lopez <klopez4212@gmail.com>
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Summary
- open Huddles in a focused companion window with a clean handoff back
to the in-app drawer and backing channel
- redesign the participant film strip, sidebar control, transcript
surface, and themed shell treatment
- preserve microphone and device control across windows, start agent
voice on the first reply, and show agent speaking activity in the film
strip
- give each agent a distinct session voice, beginning with the
configured default, plus compact per-agent text-to-speech and voice
controls
- enroll only agents explicitly mentioned or deliberately added through
an agent panel into the live Huddle roster
- keep temporary Huddle channels out of the sidebar unless the user
explicitly brings one into the main app
- remove Huddle-only avatar policy badges and filter short silence or
noise segments before speech-to-text posts
## Why
The previous flow exposed the temporary channel as product UI, obscured
who was present or speaking, and split transcript and audio state
between the main and companion windows. This keeps backing channels as
implementation details unless a user explicitly brings a Huddle into the
app, while sharing the live conversation and audio lifecycle across both
surfaces. Agent participants now join only after an explicit invitation,
distinct voices make multi-agent Huddles easier to follow, and short
microphone noise no longer becomes stray transcript messages.
## Validation
- `pnpm check`
- `pnpm build:e2e`
- `pnpm exec playwright test tests/e2e/huddle-transcription.spec.ts
--project=smoke` (13 passed)
- Huddle sidebar visibility unit coverage (4 passed)
- focused managed-agent and persona-mention E2E coverage (2 passed)
- `pnpm test` (3,910 passed)
- `cargo clippy --manifest-path desktop/src-tauri/Cargo.toml
--all-targets -- -D warnings`
- `cargo test --manifest-path desktop/src-tauri/Cargo.toml` (2,093
passed, 14 ignored; 3 diagnostics passed)
---------
Signed-off-by: kenny lopez <klopez4212@gmail.com>
Signed-off-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Wes <wesbillman@users.noreply.github.com>
Co-authored-by: Carl <c7ebe626f000404285d3686e1dc74cc07cc60a9754a150041ba132e14bd3e2ec@buzz.block.builderlab.xyz>
## Context
Pocket TTS currently offers bundled reference voices. People also need a
local, private way to add a voice without sending audio to a cloud
service.
## Summary
Add a Pocket voice import flow to Voice settings. Buzz opens the native
file picker, decodes common audio formats in the reusable `buzz-voice`
crate, canonicalizes the selected audio, stores it under a
content-derived identity in app data, selects it, and lets the user
delete it later.
## Changes
- Accept WAV, M4A, MP3, FLAC, OGG, and AIFF files between 2 and 30
seconds, including multichannel sources.
- Decode and downmix accepted audio to canonical mono 32 kHz PCM16 WAV
before hashing and storage.
- Store imported voices behind stable `pocket:imported:<sha256>`
identities and content-addressed files.
- Keep absolute file paths inside the native process and expose only
voice metadata to React.
- Include imported voices in Pocket preview and live huddle playback.
- Add Add voice and delete controls while preserving the bundled Pocket
voice catalog.
- Fall back to Mary when the selected imported voice is deleted.
- Keep durable import, selection, and deletion successful when a live
TTS worker acknowledgement is delayed.
- Preserve bundled voices when optional import metadata is unreadable
and keep failed deletion retryable.
## Related issue
None found.
## Testing
Production decoding was exercised with WAV, M4A with AAC, MP3, FLAC, OGG
Vorbis, and AIFF fixtures. Each format canonicalized to mono 32 kHz
PCM16 WAV. Manual validation in the combined daily-driver build covered
native-picker import, Preview, live-huddle playback, deletion, and Mary
fallback.
## Screenshots
The Voice settings card preserves the bundled Pocket catalog and adds
the local Add voice action.
## Reviewer-reproducible examples
Create common-format fixtures and run them through the production
importer:
```bash
. ./bin/activate-hermit
fixtures="$(mktemp -d)"
ffmpeg -hide_banner -loglevel error -f lavfi -i "sine=frequency=220:duration=3" -ac 2 -ar 44100 "$fixtures/voice.wav"
ffmpeg -hide_banner -loglevel error -i "$fixtures/voice.wav" -c:a aac "$fixtures/voice.m4a"
ffmpeg -hide_banner -loglevel error -i "$fixtures/voice.wav" "$fixtures/voice.mp3"
ffmpeg -hide_banner -loglevel error -i "$fixtures/voice.wav" "$fixtures/voice.flac"
ffmpeg -hide_banner -loglevel error -i "$fixtures/voice.wav" -c:a libvorbis "$fixtures/voice.ogg"
ffmpeg -hide_banner -loglevel error -i "$fixtures/voice.wav" -c:a pcm_s16be "$fixtures/voice.aiff"
BUZZ_VOICE_IMPORT_TEST_DIR="$fixtures" \
cargo test -p buzz-voice imports_common_audio_format_fixtures -- --ignored --nocapture
```
Exercise import persistence, synthesis, deletion, and bundled-voice
fallback with an installed Pocket model:
```bash
BUZZ_POCKET_MODEL_DIR=/path/to/pocket-model-bundle \
cargo test -p buzz-voice --test pocket_import_audio \
objective_import_synthesis_delete_and_mary_fallback \
-- --ignored --nocapture
```
Exercise the native-picker boundary, selection, preview dispatch,
deletion, cancellation, and invalid-file states:
```bash
cd desktop
pnpm build:e2e
pnpm exec playwright test tests/e2e/voice-settings.spec.ts --project=smoke
```
---------
Signed-off-by: John Tennant <jtennant@block.xyz>
Signed-off-by: John Tennant <johnmatthewtennant@gmail.com>
Signed-off-by: John Tennant <jtennant@squareup.com>
Signed-off-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Signed-off-by: npub12gtutshhh76rx0jx697f32f9tffd4hhp3hx58fp4x6u4uemkm7sqf8f757 <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Co-authored-by: John Tennant <jtennant@block.xyz>
Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: npub12gtutshhh76rx0jx697f32f9tffd4hhp3hx58fp4x6u4uemkm7sqf8f757 <5217c5c2f7bfb4333e46d17c98a9255a52dadee18dcd43a43536b95e6776dfa0@buzz.block.builderlab.xyz>
Relands **#2467** (extract `buzz-voice` crate) and **#3208** (Pocket
voice settings) onto main, after #3266 and #3180 merged.
## Why a fresh PR
The repo is squash-only with delete-branch-on-merge. Squashing #3266
deleted `jtennant/pocket-tts-2026-04`, which was #2467's base — GitHub
auto-closed #2467 and it cannot be reopened. Squash merges also sever
ancestry, so GitHub's natural merge-base reports phantom conflicts for
the whole remaining stack.
## Content provenance
- Byte-identical to the blessed `jt/buzz-voice-refactor` branch
(`93029c577`, tree `6729e0eff` — reviewed by Dawn (#2467) and Max
(#3208) at exact heads) **except** the three files where #3180 and #3208
genuinely interact.
- Three-file resolution (union of both sides):
- `huddle/mod.rs` — #3180's pipeline re-exports + #3208's
`agent_tts_routing` imports.
- `huddle/state.rs` — `reset_preserving_generation` preserves both
`huddle_generation` (#3180) and `tts_enabled` (#3208); test sets merged
into one `tests` module.
- `desktop/src/testing/e2eBridge.ts` — both switch arms kept; no
duplicate case labels.
## Verification at cf32dacd7
- `cargo test` (desktop/src-tauri, pinned 1.95.0): **2047 + 3 pass / 0
fail** (14 ignored: 8 keychain, 4 real_relay, 2 flag-gated)
- `cargo clippy --all-targets -- -D warnings`: clean; `cargo fmt
--check`: clean
- `cargo check --workspace` (root, includes new `buzz-voice` member):
clean; `cargo test -p buzz-voice`: 5/0
- `pnpm test`: **3885 / 0**; `tsc --noEmit`: clean; lint: clean
The 3180×3208 interaction resolution is getting an independent team
re-review before merge.
Buzz channel: buzz-desktop-voice `fd5fb402-b651-4238-89b1-bb3e2fa4dc96`,
thread `b4798ecc`.
Signed-off-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
Co-authored-by: npub1qyvc0c5kl4gqv2fd97fsk46tu378sqgy35vc83rvgfwne90sel7s0ed67d <011987e296fd5006292d2f930b574be47c7801048d1983c46c425d3c95f0cffd@buzz.block.builderlab.xyz>
## Context
Before this change, every huddle initialized with transcription off.
Joining or adding an agent did not enable it, so the agent could not
receive spoken conversation until a person clicked the transcript
control. Starting a huddle from an agent DM could also omit that agent,
and adding an agent who already belonged to the parent channel could
attempt an unnecessary role change and show a warning.
Agent detection uses authoritative huddle membership. A participant
counts as an agent when the ephemeral membership identifies it with the
`bot` role, or when the existing agent identity model identifies the
participant in an agent DM.
## Summary
Buzz now enables transcription once when the first authoritative agent
is present. After that initial automatic action, explicit user control
is authoritative: manual ON or OFF survives membership refreshes,
reconnects, and UI remounts. Removing the last agent does not change the
current transcription state.
Agent-DM huddles enroll the agent automatically. Adding an agent who
already belongs to the parent channel preserves the existing parent role
and completes without a role-mutation warning.
| Scenario | Before | With this change |
| --- | --- | --- |
| First authoritative agent joins or is hydrated | Transcription stays
off | Transcription turns on once |
| User explicitly turns transcription on or off | Manual control exists
without an agent policy | The explicit choice suppresses later automatic
changes |
| Last agent leaves | No defined agent-presence behavior | The current
transcription state remains unchanged |
| Huddle starts from an agent DM | The agent can be omitted | The known
agent is enrolled automatically |
| Added agent already belongs to the parent channel | Buzz can attempt a
role rewrite and warn | Existing parent membership and role are
preserved |
| Transcription is active | The control is not visually distinct | The
control is highlighted and exposes `aria-pressed=true` |
## Changes
- Derive agent presence from authoritative bot-role huddle membership
and known agent-DM identity.
- Apply the one-time auto-enable rule during create, join, membership
hydration, reconnect, pipeline startup, and local agent addition.
- Preserve explicit user state and use huddle-generation guards so stale
asynchronous work cannot alter a replacement huddle.
- Keep backend and React transcription state synchronized, with a
visible and accessible active control.
- Enroll known agent-DM participants and make parent-channel membership
updates idempotent.
- Cover hydration ordering, reconnects, remounts, explicit OFF,
last-agent removal, DM enrollment, existing membership, and active
styling.
## Related issue
None found.
## Testing
Manual validation in `pending-seed` confirmed the product contract:
1. Started a huddle from the owned, running Fizz agent DM.
2. Confirmed the authoritative roster contained the human and Fizz as an
agent.
3. Confirmed transcription enabled without clicking the control: `Stop
transcript`, `aria-pressed=true`, with the highlighted active
background.
4. Turned transcription off and confirmed `Start transcript`,
`aria-pressed=false` remained stable.
5. Removed Fizz while transcription was off and confirmed the state
stayed off.
6. Left the huddle cleanly.
## Screenshots
The same control has distinct active and inactive states.
## Reviewer-reproducible examples
From a fresh checkout:
```bash
pnpm --dir desktop build:e2e
pnpm --dir desktop exec playwright test tests/e2e/huddle-transcription.spec.ts --project=smoke
pnpm --dir desktop exec playwright test tests/e2e/mentions.spec.ts --project=smoke --grep "system agent profile exposes owned agent actions|system agent avatar exposes owned agent actions|owned bot profile exposes message and huddle actions|owned agent mention profile exposes message and huddle actions"
```
The huddle scenario exercises initial authoritative hydration, exactly
one automatic enable, explicit OFF persistence, unchanged state after
last-agent removal, newer events winning over delayed hydration,
agent-DM enrollment, and idempotent parent membership. It also asserts
`aria-pressed` and distinct computed active styling.
---------
Signed-off-by: John Tennant <jtennant@squareup.com>
Signed-off-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
Co-authored-by: npub1jh9wn95s0472h86ahapupaf7m6kx4v9sx2n0atj2hltcfer8k06s5n3pyf <95cae996907d7cab9f5dbf43c0f53edeac6ab0b032a6feae4abfd784e467b3f5@buzz.block.builderlab.xyz>
## Context
Buzz Desktop currently installs an older Pocket TTS model bundle. The
current
bundle changes the tokenizer, learned BOS input, recurrent-state
contract, and
prompt behavior, so updating download URLs alone is not compatible.
## Summary
This PR upgrades Buzz Desktop to the current pinned Pocket TTS model. It
preserves existing product behavior and the hard 50-token model-input
limit
while adding the required runtime support, verified acquisition, and
crash-safe
cache migration.
## Changes
- Pins an immutable Pocket TTS revision, artifact names, exact byte
sizes,
SHA-256 checksums, Mary reference voice, and license.
- Loads the bundle-matched SentencePiece tokenizer, learned BOS
embedding, and
bundle-declared recurrent states.
- Uses one pinned Pocket TTS configuration; no precision or
model-version
selector is added.
- Preserves the resident engine's exact `<= 50` token contract without
changing
Desktop segmentation policy.
- Bumps the Pocket cache manifest to v4, verifies size and checksum
before
adoption, atomically swaps the cache, and recovers the last verified
cache
after interrupted installs, including an incomplete final directory.
- Keeps acquisition, cache migration, worker adoption, and tests within
the
existing Desktop implementation.
- Removes the obsolete model-quality harness, which was coupled to the
superseded production prompt and model layout.
## Related issue
None.
## Testing
Manual listening completed on the exact Desktop build. The updated model
improved speech quality and resolved the phrase-start and sample-onset
artifacts. Reproducible integrity and model checks are below.
## Screenshots
N/A. This changes model installation and speech synthesis, not a visual
surface.
## Reviewer-reproducible examples
### Before and after model identity
```sh
git show 35305bfc8fd456ca9a17caa1ddbfaabd87d46981:desktop/src-tauri/src/huddle/models.rs \
| grep -E 'sherpa-onnx-pocket-tts|TTS_MODEL_VERSION'
git show 211d17c58567448fe7ac95c4fa0ad2b88378849a:desktop/src-tauri/src/huddle/pocket_models.rs \
| grep -E 'MODEL_REPOSITORY|MODEL_REVISION|MODEL_PRECISION|MAX_TOKENS'
```
The target branch identifies the January bundle. The PR branch
identifies the
immutable April revision, INT8 precision, and 50-token maximum.
### Deterministic runtime validation
Use the pinned artifacts listed in `pocket_models.rs` and run the
model-dependent Pocket tests with the model directory supplied by the
test environment. The checked-in long-sentence fixture must preserve its
expected 48 and 44 token split and produce non-silent PCM.
### Manual listening validation
John listened to an untrimmed Pocket TTS onset-stress clip generated
from the exact user-provided passage, with every sentence synthesized
separately and identical 100 ms digital-silence boundaries. The clip
used no leading period, onset trimming, gain adjustment, or loudness
normalization.
The updated model produced better-quality speech and resolved the
start-of-sample artifacts.
---------
Signed-off-by: John Tennant <jtennant@block.xyz>
Signed-off-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
Co-authored-by: John Tennant <jtennant@block.xyz>
Co-authored-by: npub1mprnacetjua2xx3p5eddmhxyk6wv929ymm5py8kd2xfxurxahspqqlgyta <d8473ee32b973aa31a21a65adddcc4b69cc2a8a4dee8121ecd51926e0cddbc02@buzz.block.builderlab.xyz>
**Category:** new-feature
**User Impact:** Users can create, download, and verify a
password-protected backup of their private identity from desktop
Settings.
**Problem:** Buzz does not currently give signed-in users a
Settings-based path to protect or validate their private identity
independently of onboarding. **Solution:** Add a focused backup menu to
the private-key row, keep encryption and verification local in Rust, and
preserve completed encrypted backups briefly so native saves can be
retried without repeating encryption.
<details>
<summary>File changes</summary>
**desktop/src/features/settings/**
Adds the background backup lifecycle, create and test dialogs,
private-key menu integration, password handling, and focused unit
coverage.
**desktop/src/features/onboarding/ui/NsecMaskedDisplay.tsx**
Extends the masked private-key display with reusable overflow-menu
actions used by Settings.
**desktop/src/app/App.tsx**
Mounts the backup provider at app scope so encryption and save work
survive closing Settings or the modal.
**desktop/src/shared/api/tauriIdentity.ts**
Adds typed desktop bindings for local backup creation, save, selection,
and verification.
**desktop/src-tauri/src/key_backup.rs and
desktop/src-tauri/src/commands/identity.rs**
Implements local NIP-49 encryption, password generation, file handling,
and public-identity-only verification results.
**desktop/src-tauri/src/egress_guard.rs and guarded call sites**
Blocks encrypted secret material from relay, websocket, snapshot,
sharing, and huddle egress paths.
**desktop/src-tauri tests and fixtures**
Covers encryption, verification, file behavior, and fail-closed
no-egress protections.
**desktop/src/testing/e2eBridge.ts, desktop/tests/, and
desktop/playwright.config.ts**
Expands the mock native bridge and browser coverage across create,
retry, expiry, and current/different-identity verification states.
**desktop/src-tauri/Cargo.toml, Cargo.lock, and assets**
Adds the local cryptography/password-generation dependencies and
embedded short-word list.
</details>
## Reproduction steps
1. Run the desktop app and open **Settings → Profile → Identity**.
2. Open the private-key overflow menu and choose **Create backup**.
3. Enter or generate a valid password, submit, and confirm progress
continues if the dialog or Settings is closed.
4. Save the resulting `.ncryptsec` file; cancel and retry to confirm the
temporary download remains available.
5. Choose **Test backup**, select the file, enter a wrong password, then
retry with the correct password.
6. Confirm success identifies whether the backup matches the current
identity and displays only the public `npub`.
## Screenshots
| Settings identity | Private-key menu | Create backup |
|---|---|---|
| <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/981e391b-6829-4081-95ca-ca75a369de71"
/> | <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/7972c68e-7635-47d8-b0ad-9639390d3e6c"
/> | <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/4709c8f7-cf02-46f1-bec9-b3f98fe56fb2"
/> |
| Encrypting | Download available | Test success |
|---|---|---|
| <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/1ac3e934-2b4b-4135-bae6-126c715c8c59"
/> | <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/cb6f07ee-a16f-44a5-b9a0-6b9fe0e4d40d"
/> | <img width="1280" height="720" alt="image"
src="https://github.com/user-attachments/assets/ea58b1b1-966c-46aa-8d59-92c9f06a25bd"
/> |
Visual review and additional states: [Buzz
thread](buzz://message?channel=50ca7ef1-201e-4159-9499-40de3964b7c3&id=87eceb5f0f82fd50c32e560de3d35be48e293760f6620718aafdcef289d475fe)
---------
Signed-off-by: Taylor Ho <taylorkmho@gmail.com>
Co-authored-by: npub1223z34hd7vtwc6qj4s7flsxkj644nlre2nthu7lrrmkumhu3xddsrx9r6w <52a228d6edf316ec6812ac3c9fc0d696ab59fc7954d77e7be31eedcddf91335b@buzz.block.builderlab.xyz>
## Summary
- Preserve plain-string errors returned by Tauri and show an actionable
message when huddle audio is unavailable in the relay deployment.
- Use one error formatter across channel, timeline, wave, and profile
huddle actions while preserving other relay and device errors.
- Complete the huddle lifecycle when audio setup fails after publishing
a start event, preventing peers and reloaded clients from reconstructing
a phantom active huddle.
- Cover unavailable-audio formatting and START → rollback-END
reconstruction with regression tests.
## Behavior
The Tauri huddle commands reject with a plain string. Several desktop
toast call sites only preserved JavaScript `Error` objects, so the relay
message was discarded and replaced with “Failed to join huddle.”
The desktop now recognizes `huddle_audio_unavailable` and the current
relay message, then shows:
> Huddle audio isn’t available on this server. Ask an administrator to
turn it on.
Other relay and device messages remain intact, including microphone
errors.
`start_huddle` also publishes `KIND_HUDDLE_STARTED` before audio setup.
If setup fails, rollback now publishes `KIND_HUDDLE_ENDED` through the
normal end-and-archive path before resetting local state. This makes the
failed start observable to lifecycle reconstruction and prevents stale
join affordances.
## Checks
- `cd desktop && pnpm test` — 3,405 passed
- `cd desktop/src-tauri && cargo test` — 1,560 passed, 13 ignored; 3
diagnostic tests passed
- `cd desktop && pnpm exec playwright test tests/e2e/channels.spec.ts
--project=smoke --grep 'huddle rollback end event'` — passed
- `just desktop-tauri-check`
- `cd desktop && pnpm typecheck`
- `cd desktop && pnpm check`
- Pre-commit and pre-push hooks
