Files
LocalAI/pkg/model/load_observer_test.go
Richard Palethorpe 5d0c43ec6e feat(realtime): Semantic VAD EOU token (#10444)
* feat(realtime): EOU-driven semantic_vad turn detection

Add a `semantic_vad` turn-detection mode to the realtime API that feeds
the transcription model live and decides "the user finished speaking"
from the `<EOU>` end-of-utterance token rather than from silence alone.
When EOU fires the turn commits immediately (~0.3s); otherwise it falls
back to an eagerness-scaled silence threshold (low/med/high = 8/4/2s).

Plumbing, bottom to top:

- proto: `AudioTranscriptionLive` bidirectional RPC (config-first oneof,
  mono float PCM @16k, ready-ack / Unimplemented degrade signal) plus
  `TranscriptResult.eou` for the unary retranscribe gate.
- pkg/grpc: client/server/base/embed scaffolding for the bidi stream,
  modeled on AudioTransformStream; release stream conns on terminal Recv.
- parakeet-cpp: live transcription RPC with per-C-call engine locking
  (one live stream per turn, finalize+free at commit); bump parakeet.cpp
  to ABI v5 — incremental StreamingMel (no more quadratic per-feed mel
  recompute that delayed EOU on long turns) and the <EOU>/<EOB> split;
  strip the literal <EOU>/<EOB> from offline text and set Eou.
- core/backend: LiveTranscriptionSession wrapper + pipeline
  `turn_detection:` config block (type/eagerness/retranscribe).
- realtime: semantic_vad integration — live input captions streamed as
  transcription deltas while the user speaks, EOU-immediate commit with
  eagerness fallback, optional retranscribe gate (batch re-decode must
  also end in <EOU> to confirm), clause synthesis off the LLM token
  callback, and per-turn live-transcription / model_load telemetry.
- UI: show the realtime pipeline components as a vertical list.

Docs and tests included; opt-in via the pipeline YAML or per-session
`session.update`. Non-streaming STT backends degrade to silence-only.

Assisted-by: Claude Code:claude-opus-4-8 [Read] [Edit] [Write] [Bash]
Assisted-by: Claude Code:claude-fable-5 [Read] [Edit] [Bash]
Signed-off-by: Richard Palethorpe <io@richiejp.com>

* feat(realtime): explicit formally-verified state machines + parakeet streaming driver

The realtime API had several implicit state machines whose state was inferred
from scattered booleans, channels, and five separate mutexes, leaving
illegal/inconsistent states reachable. Make them explicit and keep the
implementation in step with a formal design; rework the parakeet streaming
backend along the same lines.

Realtime state machines (M1-M5). Each is a sealed sum-type State/Event/Effect
with a total, pure Next(state,event)->(state,[]effect) behind a single-writer
Coordinator:

  M1 conncoord    connection lifecycle: VAD toggle + once-only teardown
                  (replaces vadServerStarted + a `done` channel closed from
                  two sites).
  M2 turncoord    turn detection: collapses speechStarted and the live-stream
                  "turn open" flag into one state, so discardTurn can no longer
                  desync them and suppress the next onset.
  M3 respcoord    response coordination: serializes the dual-writer
                  start/cancel so at most one response is live; one
                  response.done per response.create.
  M4 compactcoord conversation compaction: single-flight (replaces the
                  `compacting atomic.Bool` CAS).
  M5 ttscoord     TTS pipeline: open->closing->closed, idempotent wait(),
                  rejects enqueue-after-close (was a silent drop).

The Coordinator/Sink/Next plumbing — only the sealed types and Next differed
per machine — is extracted once into core/http/endpoints/openai/coordinator as
a generic Coordinator[S,E,F]; each machine keeps its public API via type
aliases, so no sink, call-site, or test moved.

Hierarchy. session_lifecycle.fizz models M1 as the parent region with its
children (M2/M3/M4) as one statechart and asserts ChildrenDieWithParent (conn
torn => all children terminal, none start after teardown). respcoord and
compactcoord gain an absorbing Terminated state + Shutdown event; conncoord's
teardown drives the children terminal. This closes a compaction teardown gap: a
fire-and-forget compaction could outlive a torn session — compactionSink now
takes a session-scoped cancellable context + WaitGroup and joins the in-flight
summarize+evict on shutdown.

Formal verification. formal-verification/ holds one authoritative FizzBee spec
per machine plus the composition spec, each with an always-assertion and a
documented one-line edit that makes the checker fail (verified non-vacuous).
scripts/realtime-conformance.sh is fail-closed: all Go conformance suites under
-race AND a model-check of every .fizz spec; a missing FizzBee is a hard error
(only the loud REALTIME_CONFORMANCE_SKIP_FIZZBEE=1 bypasses it, never in CI).
FizzBee is pinned by sha256 and installed via scripts/install-fizzbee.sh into
.tools/ (gitignored). Wired as make test-realtime-conformance, a CI workflow,
and a pre-commit path filter. Go conformance tests are Ginkgo/Gomega (per the
repo's forbidigo lint): transition tables + fixed-seed property walks +
concurrent/-race specs, no rapid dependency. Design map:
docs/design/realtime-state-machines.md.

Parakeet streaming backend. The same treatment applied to the parakeet-cpp
streaming paths:
- AudioTranscriptionStream returns codes.Unimplemented for non-streaming models
  instead of decoding offline and emitting it as one delta + final. A client
  that asked for streaming learns the model cannot stream rather than receiving
  a batch result shaped like a stream. New grpcerrors.StreamTranscriptionUnsupported
  carries that signal; the HTTP /v1/audio/transcriptions stream path surfaces it
  as an SSE error event. Mirrors AudioTranscriptionLive, which already did this.
- utteranceBoundary (boundary.go): a single definition of the end-of-utterance
  latch, replacing three open-coded finalEou toggles. Modelled as a two-valued
  type so illegal states are unrepresentable.
- Shared decode driver (driver.go): streamFeedResult (one per-feed event) +
  feedChunk (hides the ABI v4 JSON vs text-only split) + feedSlices + flushTail.
  The feed loop is written once.
- AudioTranscriptionLive becomes a bidi adapter: it streams the per-feed
  {delta,eou,eob,words} the realtime turn detector consumes and a terminal
  FinalResult carrying only Text. Segments/duration/eou are offline-only and no
  longer produced (nor read) on the live path; liveTraceState drops the terminal
  eou and keeps the per-feed eou_events count.
- AudioTranscriptionStream + streamJSON merge into one driver-based function;
  streamSegmenter is generalized to the unified event with a text-only fallback
  that preserves the legacy (no-words) library's per-utterance segmentation.

Verified: build/vet/gofumpt clean, golangci-lint 0 issues, all coordinator and
parakeet packages under -race, the fail-closed conformance gate green, and
make test-realtime (12 e2e WS+WebRTC).

Assisted-by: Claude:claude-opus-4-8 [Claude Code]
Signed-off-by: Richard Palethorpe <io@richiejp.com>

---------

Signed-off-by: Richard Palethorpe <io@richiejp.com>
2026-06-30 09:01:22 +02:00

95 lines
2.7 KiB
Go

package model_test
import (
"os"
"github.com/mudler/LocalAI/pkg/model"
"github.com/mudler/LocalAI/pkg/system"
. "github.com/onsi/ginkgo/v2"
. "github.com/onsi/gomega"
)
// The load observer is the hook the core wires to backend tracing: it must
// fire once per actual load attempt with the alias-resolved backend and the
// resolved runtime URI, because that pair is what tells an operator WHICH
// installed backend build served a model (a stale install is invisible in
// the model config but shows up here).
var _ = Describe("ModelLoader load observer", func() {
var (
modelLoader *model.ModelLoader
modelPath string
events []model.BackendLoadEvent
)
BeforeEach(func() {
var err error
modelPath, err = os.MkdirTemp("", "load_observer")
Expect(err).ToNot(HaveOccurred())
systemState, err := system.GetSystemState(
system.WithModelPath(modelPath),
)
Expect(err).ToNot(HaveOccurred())
modelLoader = model.NewModelLoader(systemState)
events = nil
modelLoader.SetLoadObserver(func(ev model.BackendLoadEvent) {
events = append(events, ev)
})
})
AfterEach(func() {
Expect(os.RemoveAll(modelPath)).To(Succeed())
})
It("fires with the resolved runtime URI when a load attempt fails", func() {
// A non-file external backend URI is treated as a remote gRPC
// address; nothing listens on port 1, so the health check fails
// after the single configured attempt — a real, fast load attempt.
modelLoader.SetExternalBackend("fakebackend", "127.0.0.1:1")
_, err := modelLoader.Load(
model.WithModelID("m"),
model.WithModel("m.bin"),
model.WithBackendString("fakebackend"),
model.WithGRPCAttempts(1),
model.WithGRPCAttemptsDelay(0),
)
Expect(err).To(HaveOccurred())
Expect(events).To(HaveLen(1))
ev := events[0]
Expect(ev.ModelID).To(Equal("m"))
Expect(ev.ModelName).To(Equal("m.bin"))
Expect(ev.Backend).To(Equal("fakebackend"))
Expect(ev.BackendURI).To(Equal("127.0.0.1:1"))
Expect(ev.Err).To(HaveOccurred())
})
It("fires for unknown backends with an empty runtime URI", func() {
_, err := modelLoader.Load(
model.WithModelID("m"),
model.WithModel("m.bin"),
model.WithBackendString("no-such-backend"),
)
Expect(err).To(HaveOccurred())
Expect(events).To(HaveLen(1))
Expect(events[0].Backend).To(Equal("no-such-backend"))
Expect(events[0].BackendURI).To(BeEmpty())
Expect(events[0].Err).To(MatchError(ContainSubstring("backend not found")))
})
It("is optional: loads proceed when no observer is registered", func() {
modelLoader.SetLoadObserver(nil)
_, err := modelLoader.Load(
model.WithModelID("m"),
model.WithModel("m.bin"),
model.WithBackendString("no-such-backend"),
)
Expect(err).To(HaveOccurred())
Expect(events).To(BeEmpty())
})
})