MCPcopy Create free account

hub / github.com/debpalash/VoiceStudio / types & classes

Types & classes1,240 in github.com/debpalash/VoiceStudio

↓ 45 callersClassJudgeResult
The atomic verdict. ``passed`` is True/False, or None when skipped (e.g. an optional backend isn't installed).
tests/probe/spec.py:25
↓ 45 callersClassWorkerError
A failure as it crosses the wire — already scrubbed, always actionable.
backend/worker/errors.py:137
↓ 33 callersClassTaskExecutor
Executes protocol assignments against the local engine stack.
backend/worker/executor.py:92
↓ 25 callersClassEvent
backend/worker/inbound/connection_log.py:48
↓ 25 callersClassScheduler
Owns the queue, the selection pipeline, and the deadline sweeper.
backend/worker/scheduler.py:154
↓ 23 callersClassTranslateRequest
backend/schemas/requests.py:147
↓ 22 callersClassConnection
tests/test_auth_principal.py:49
↓ 22 callersClassWorkerPool
The set of workers currently connected, plus their breakers.
backend/worker/pool.py:146
↓ 20 callersClassFakePlane
tests/test_gpu_gateway.py:97
↓ 20 callersClass_Plane
A control plane stub with a real pool, so availability is real logic.
tests/test_worker_routing.py:41
↓ 17 callersClassExpressiveOptions
Optional expressive/quality knobs for a longform render (#1208). Every field is ``None``/``False`` by default, and a default instance means *
backend/services/audiobook.py:69
↓ 17 callersClassFakeScheduler
Enough scheduler to exercise submit / wait / cancel / get.
tests/test_gpu_gateway.py:36
↓ 17 callersClassSpan
One contiguous run of text in a single voice, plus trailing silence. ``speed`` (when set) is the per-span rate passed to the engine — Stories'
backend/services/audiobook.py:174
↓ 17 callersClassTranslateSegment
backend/schemas/requests.py:128
↓ 17 callersClass_Context
tests/test_worker_upload_server.py:80
↓ 16 callersClassHostCaps
Snapshot of the host's accelerator capability. Immutable + cached.
backend/core/device_caps.py:401
↓ 15 callersClassChapter
backend/services/audiobook.py:192
↓ 15 callersClassContext
tests/test_worker_inbound_transport.py:681
↓ 15 callersClassDubRequest
backend/schemas/requests.py:58
↓ 15 callersClassFakeGateway
Stands in for the scheduler half of `services/gpu_gateway.py`. Same signature and same return contract as the real `run` — a `(waveform, samp
tests/test_generation_remote_routing.py:119
↓ 14 callersClass_FakeAssignment
Only the attributes the executor reads — a real pb message needs a ref and a session the executor has no business knowing about.
tests/test_worker_executor_residency.py:35
↓ 13 callersClassAdminSessionStore
Thread-safe, process-local store for admin sessions and WS tickets.
backend/services/admin_sessions.py:100
↓ 13 callersClass_Wire
Drains the client's outbox, recording each frame's arrival time.
tests/test_worker_client_liveness.py:82
↓ 12 callersClassResponse
backend/tests/test_community.py:221
↓ 12 callersClassSegment
backend/services/segmentation.py:76
↓ 12 callersClassTaskFailure
Carries a classified ``WorkerError`` across the execution boundary.
backend/worker/executor.py:841
↓ 12 callersClassWorkerConfig
Everything the worker needs to reach and prove itself to a server.
backend/worker/transport/client.py:257
↓ 12 callersClass_ScriptedLLMClient
OpenAI-compatible fake. `script(call_kwargs) -> str | Exception` decides each response; every chat.completions.create() call is recorded.
tests/test_dub_translation_quality.py:14
↓ 11 callersClassConnection
backend/worker/inbound/connection_string.py:39
↓ 11 callersClassConnection
tests/test_worker_server_integrity.py:535
↓ 11 callersClassFitParams
Tunable knobs for the Smart Fit planner. All defaults are deliberately conservative: ≤1.2× audio-only is imperceptible to most listeners; 1.5
backend/services/fit_planner.py:62
↓ 11 callersClassWorkerClient
Maintains one worker's connection to a control plane.
backend/worker/transport/client.py:286
↓ 11 callersClassWorkerServicer
Implements ``WorkerService`` on top of the scheduler and registry.
backend/worker/transport/server.py:656
↓ 11 callersClass_FakeResponse
tests/test_media_tools.py:70
↓ 11 callersClass_StubModel
tests/test_clone_prompt_cache.py:14
↓ 11 callersClass_Wire
Drains the client's outbox, recording each frame's arrival time.
tests/test_worker_upload_client.py:280
↓ 10 callersClassFakePage
Minimal stand-in for a Playwright sync Page — enough for the judges + the Driver's locate logic to run with no browser.
tests/probe/test_probe_web.py:19
↓ 10 callersClassNlmsEchoCanceller
Time-domain NLMS adaptive filter with Geigel double-talk detection. Operates on narrowband mono PCM at 16 kHz (the rate the dictation path re
backend/services/aec.py:75
↓ 10 callersClass_FakeStub
Records what UploadResult received, and can be told to refuse.
tests/test_worker_upload_client.py:53
↓ 10 callersClass_SafetensorError
Stands in for safetensors_rust.SafetensorError — the point being that it is NOT an OSError, which is why the ladder never saw the real one.
tests/test_corrupt_weights_recovery_1406.py:97
↓ 9 callersClassDubSegment
backend/schemas/requests.py:19
↓ 9 callersClassEvalCase
A complete evaluation scenario.
tests/evals/case.py:30
↓ 9 callersClassLLMJudge
Scores case transcripts against a rubric via the configured LLM.
tests/evals/judge.py:68
↓ 9 callersClassTask
A unit of inference work, independent of which worker runs it.
backend/worker/lifecycle.py:289
↓ 9 callersClass_FakeModel
Counts encodes so cache hits are observable.
tests/test_pockettts_sidecar.py:201
↓ 8 callersClassDecision
Where one job should run, and why.
backend/worker/routing.py:239
↓ 8 callersClassOmniVoiceSubprocessBackend
The resident OmniVoice model in a killable sidecar process.
backend/engines/omnivoice_subprocess/__init__.py:45
↓ 8 callersClass_Engine
tests/test_engine_language_rejection.py:28
↓ 8 callersClass_Scheduler
The endpoint's whole contract with the scheduler: submit, wait, cancel.
tests/test_worker_service_api.py:631
↓ 8 callersClass_StepError
Install-step failure carrying user-facing remediation text.
backend/services/sidecar_install.py:378
↓ 8 callersClass_StubModel
Counts reference encodes vs. generate calls — the whole point of the test. Mirrors the real model's contract: ``voice_clone_prompt`` and ``re
tests/test_clone_prompt_wiring.py:43
↓ 7 callersClassArtifactStore
Node-side staging for one listener. Shared across panels.
backend/worker/inbound/artifacts.py:93
↓ 7 callersClassClient
backend/tests/test_community.py:228
↓ 7 callersClassFakeJudgeBackend
Canned judge transport: returns queued raw strings.
tests/evals/test_evals_harness.py:23
↓ 7 callersClassKeyStore
Thread-safe, file-backed store of per-panel keys. Backed by a plain JSON file rather than the settings store because the settings store is re
backend/worker/inbound/keys.py:140
↓ 7 callersClassModelSlot
Capacity bookkeeping for one (worker, model) pair.
backend/worker/capacity.py:119
↓ 7 callersClassRuleDurationEstimator
omnivoice/utils/duration.py:31
↓ 7 callersClassStub
tests/test_worker_inbound_transport.py:551
↓ 7 callersClass_FakeBackend
tests/backend/services/test_refinement_llm.py:54
↓ 7 callersClass_Info
tests/test_dub_batch_width_and_budget.py:121
↓ 7 callersClass_StubModel
tests/test_prompt_disk_cache.py:44
↓ 6 callersClassAttempt
One try of a task on one worker.
backend/worker/lifecycle.py:197
↓ 6 callersClassFakeTranscriber
Deterministic transcriber for the harness's own offline tests. Either returns a fixed string, or echoes a per-path mapping. Lets us prove the
tests/probe/judges/transcription.py:85
↓ 6 callersClassGalleryError
The gallery answered, and what it said cannot be trusted or used.
backend/services/gallery.py:114
↓ 6 callersClassLifecycleError
An illegal transition was attempted.
backend/worker/lifecycle.py:179
↓ 6 callersClassRefinementFlags
Which refinement behaviours to apply.
backend/services/refinement.py:176
↓ 6 callersClassRemoteJobFailed
Remote work started and then failed. Rule 2: this is not a fallback. Carries what a caller needs to offer "Run locally instead" — the same re
backend/services/gpu_gateway.py:120
↓ 6 callersClassSentenceChunker
Accumulates streaming tokens and yields complete sentences. Usage:: chunker = SentenceChunker() for token in llm_stream:
backend/services/sentence_chunker.py:303
↓ 6 callersClassTaskManager
In-memory task dispatcher with SQLite-backed metadata. The dispatcher itself (queue + worker + listeners) stays in-memory for speed, but ever
backend/core/tasks.py:13
↓ 6 callersClassTerminalRegistrationError
A registration failure that reconnecting cannot repair.
backend/worker/transport/client.py:141
↓ 6 callersClass_CountingFactory
Stands in for get_active_asr_backend; counts backend constructions — the expensive per-call model load the cache must eliminate.
tests/test_transcribe_reference.py:105
↓ 6 callersClass_Engine
tests/test_dub_batch_width_and_budget.py:27
↓ 6 callersClass_FakeASR
Minimal ASR backend; `turns` simulates FunASR-style inline diarization.
tests/test_speaker_hint.py:53
↓ 6 callersClass_FakeBackend
Stands in for a warm mlx-whisper / sherpa recognizer.
tests/test_capture_asr_idle_release.py:30
↓ 6 callersClass_FakeCuda
tests/test_torch_compile_gate.py:25
↓ 6 callersClass_FakeReq
tests/test_dub_translate.py:49
↓ 6 callersClass_FakeSeg
tests/test_dub_translate.py:42
↓ 5 callersClassBreakerRegistry
All breakers for all workers, plus fleet-wide event detection. Session-scoped by design: OSS control planes restart constantly (the app quits
backend/worker/breaker.py:213
↓ 5 callersClassControlContext
tests/test_worker_upload_server.py:294
↓ 5 callersClassEchoBackend
In-test subclass — runs the echo sidecar under `sys.executable`.
tests/backend/services/test_subprocess_backend.py:41
↓ 5 callersClassEvalRunner
tests/evals/runner.py:44
↓ 5 callersClassEvalTurn
A single user utterance in a scripted conversation.
tests/evals/case.py:20
↓ 5 callersClassFakePool
tests/test_gpu_gateway.py:104
↓ 5 callersClassInvalidBinaryError
A managed executable failed pre-exec validation (missing, empty, or not a real executable), or the OS refused to exec it. The message is user-
backend/services/binary_preflight.py:25
↓ 5 callersClassRemoteWorker
A worker as the control plane knows it between connections.
backend/worker/registry.py:50
↓ 5 callersClass_Fake
tests/test_llm_skills.py:244
↓ 5 callersClass_FakeASR
Stands in for the active ASR backend's .transcribe() — no model, no network. `chunks_by_path` maps a ref_audio path to the canned chunk list t
tests/test_speaker_clone_purity.py:175
↓ 5 callersClass_FakeBackend
tests/test_transcribe_reference.py:26
↓ 5 callersClass_FakeChatter
Non-Off LLM stand-in that always returns one fixed reply; counts calls.
tests/test_autofit_838.py:21
↓ 5 callersClass_FakeDiarization
Mimics pyannote's `Annotation.itertracks(yield_label=True)`.
tests/test_assign_speakers_from_diarization.py:20
↓ 5 callersClass_FakeModel
Model whose ``generate`` raises the given exceptions, in order, then succeeds.
tests/test_compile_fallback.py:87
↓ 4 callersClassAudiobookPreviewRequest
backend/api/routers/audiobook.py:773
↓ 4 callersClassAudiobookRequest
backend/api/routers/audiobook.py:217
↓ 4 callersClassBackend
tests/test_gpu_gateway.py:268
↓ 4 callersClassContext
tests/test_worker_server_integrity.py:1408
↓ 4 callersClassFakeClient
tests/test_worker_inbound_transport.py:659
↓ 4 callersClassInboundConnectionError
A pasted inbound connection could not be validated or activated.
backend/worker/inbound/connector.py:67
↓ 4 callersClassInboundConnectionRollbackError
A failed connection change could not restore its prior generation.
backend/worker/inbound/connector.py:71
↓ 4 callersClassIndexTTS2Backend
IndexTTS 2.5 (Bilibili) — isolated subprocess with IndexTTS-2 fallback. Plan 02-03 migrated IndexTTS off the in-process import path because I
backend/engines/indextts/__init__.py:89
↓ 4 callersClassNodeConnection
One panel→node session, reconnecting until told to stop.
backend/worker/inbound/connector.py:151
next →1–100 of 1,240, ranked by callers