Move STT and TTS to the workstation, where the microphone already is #209

2026-08-08T22:47:07+02:00

claude commented

2026-08-08 22:47:07 +02:00

Closes Vikunja #486.

Acceptance criteria and quality gate are in TASK.md on this branch.
Review the review, not the diff — leave comments and the agent will apply them via task start 486.

Closes Vikunja #486. Acceptance criteria and quality gate are in `TASK.md` on this branch. Review the review, not the diff — leave comments and the agent will apply them via `task start 486`.

claude added 2 commits 2026-08-08 22:47:07 +02:00

Point this box at the workstation transcriber (V-486) 944e553669

The block is inert until the code in PR #208 lands, and deleting it sends
every utterance back to mavsttd, which is what the box does today.

Port 8081 and not mavgpud's 8080, because whisper.cpp cannot load
CrisperWhisper 2.0 at all and it runs under transformers as its own service.
The token comes from deploy/telegram.env like every other secret here. It is
what stops anything on the LAN posting audio to that port.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

Record which machine hears him now (V-486) 4b1edb0617

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

kami added 3 commits 2026-08-08 23:08:05 +02:00

One owner for the card, not two neighbours (V-486) b975716759

CW2 is a ROCm process, so it registers on the KFD like any contender. Running
it as its own systemd unit made mavgpud yield llama-server to it every few
seconds. The gemma-4-12b arm was down for eight minutes on 2026-08-09 and
routing had silently fallen back to the resident model.

So mavgpud takes an `stt` block and runs the transcriber itself. `foreign` now
excludes every child rather than one pid, which is the fix. Yielding is all or
nothing, because a job that wants the card wants all of it. Idle unloading
stays llama-server's alone: CW2 holds 1.6GB and unloading it would only send
the next voice turn to the homesrv floor.

Maven still talks to the transcriber directly on 8081. There is no proxy,
because with no idle timer there is nothing for one to measure.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

The transcriber ships with the daemon that starts it (V-486) 1456336652

serve.py lived only on workpc, which was fine while systemd launched it and is
not fine now that mavgpud does. Two endpoints and no framework: /health answers
503 until the model is loaded, /transcribe takes raw PCM and returns
{"text","confidence"}.

The unit carries CW2_TOKEN through EnvironmentFile and the child inherits it,
so the token is never a flag value.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

Say that the card takes one supervisor (V-486) 22a4978459

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

kami added 1 commit 2026-08-08 23:15:02 +02:00

Swap the workstation model to gemma-4-E4B (V-486) a1a2fa3704

Owner's call. E4B is 4.2GB against 6.7GB plus a 0.86GB draft, so with CW2
resident the card holds 5.8GB of 16GB instead of 9.2GB.

Measured against a same-session 12B control on the 96-case fixture: 83.3% full
against 84.4%, 89.6% intent-only against 91.7%, destination 19/33 against
23/33, p50 294ms against 344ms. Destination is the column that moved. E4B names
nothing where the 12B names recall or calendar, which walks the whole chain
rather than answering wrong.

MTP is gone with the 12B and cannot come back. It is a separate gguf of
architecture gemma4-assistant with nextn_predict_layers=4, and the only one on
disk is trained against the 12B's hidden states. Neither target gguf carries
nextn tensors, so neither self-speculates.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_013ptwopxyo3Z2kwFckHkLvN

kami added 1 commit 2026-08-08 23:26:23 +02:00

Merge remote-tracking branch 'origin/master' into task/486-deploy-the-workstation-transcriber 8aba4845bf

claude merged commit 6d5801bb1f into master

2026-08-08 23:26:31 +02:00

Sign in to join this conversation.

2 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: kami/Maven#209