Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/debpalash/VoiceStudio
/ functions
Functions
15,058 in github.com/debpalash/VoiceStudio
⨍
Functions
15,058
◇
Types & classes
1,240
↳
Endpoints
303
↓ 1 callers
Function
_coerce_int
(value, default: int = 0)
backend/api/routers/longform_jobs.py:55
↓ 1 callers
Function
_collapse_character_runs
(text: str, min_run: int)
backend/services/refinement.py:154
↓ 1 callers
Function
_collapse_word_runs
(text: str, min_run: int)
backend/services/refinement.py:125
↓ 1 callers
Function
_collect
()
tests/test_dub_no_tts_load_for_asr.py:101
↓ 1 callers
Function
_collect
()
tests/test_funasr_global_speaker_chunking.py:133
↓ 1 callers
Function
_collect
()
tests/test_tts_placement_self_heal.py:205
↓ 1 callers
Function
_collect
()
tests/test_speaker_hint.py:131
↓ 1 callers
Function
_combine_text
(text, ref_text: Optional[str] = None)
omnivoice/models/omnivoice.py:1770
↓ 1 callers
Function
_commit
Finalize one utterance: decode it off-thread, flush a polished `final`, drop its samples from the live buffer. `receive()` may append
backend/api/routers/capture_ws.py:991
↓ 1 callers
Function
_commit_touches_inference
Heuristic: does this commit's tree contain ONNX weights / tokenizer? A commit that only adds a README or sample audio doesn't change inferenc
scripts/resolve_supertonic3_sha.py:67
↓ 1 callers
Function
_community_audio_is_current
(row, item: dict, ref_text: str)
backend/api/routers/community.py:683
↓ 1 callers
Function
_compile
Build the single alternation regex + a casefold→respelling lookup. Keys are sorted longest-first so an overlapping longer key (``Dr. Smith``)
backend/services/pronunciation.py:72
↓ 1 callers
Function
_compile_abbreviations
()
backend/services/text_normalization.py:332
↓ 1 callers
Function
_compile_phrase
Compile a synonym phrase to a regex. Latin phrases get word boundaries (so "male" never fires inside "female", "old" never inside "bold") and
backend/core/describe_voice.py:199
↓ 1 callers
Function
_compiled_rules
()
backend/core/describe_voice.py:213
↓ 1 callers
Method
_complete_registration
Validate durable acceptance, then run the exact issued session.
backend/worker/inbound/connector.py:410
↓ 1 callers
Function
_compute_status
()
backend/core/cudnn8.py:190
↓ 1 callers
Function
_concat_escape
Escape a path for a concat-demuxer list `file '<path>'` directive.
backend/services/video_retime.py:223
↓ 1 callers
Function
_concat_slices
Concatenate the picked segment audio slices into one reference array.
backend/services/speaker_clone.py:377
↓ 1 callers
Function
_config_path
Locate Confucius4's inference config (``config/inference_config.yaml``) under the clone, or an explicit override.
backend/engines/confucius4/main.py:82
↓ 1 callers
Function
_configure_rocm_if_needed
Auto-set HSA_OVERRIDE_GFX_VERSION for AMD GPUs on ROCm. ROCm-enabled PyTorch reports `torch.cuda.is_available() == True` but some consumer AM
backend/services/model_manager.py:1321
↓ 1 callers
Function
_configured_pin
(connection)
backend/core/auth.py:284
↓ 1 callers
Function
_configured_pin
The active share PIN (``app.state.network_share.pin``) or None. Read via getattr so a bare Request stub (or a request that hit before lifespan set
backend/api/dependencies.py:66
↓ 1 callers
Method
_connect_once
(self)
backend/worker/transport/client.py:446
↓ 1 callers
Method
_connect_once
(self)
tests/test_worker_agent.py:1044
↓ 1 callers
Function
_consent_verified
B12-B16: trust verified-own-voice ONLY with a real recording (≥ floor) AND non-empty consent_text AND a consent.json present. The manifest flag al
backend/api/routers/personas.py:124
↓ 1 callers
Function
_control
(args: argparse.Namespace, action: str)
backend/speech_client/__main__.py:132
↓ 1 callers
Function
_convolve1d
(signal: torch.Tensor, kernel: torch.Tensor)
omnivoice/scripts/extract_audio_tokens_add_noise.py:265
↓ 1 callers
Function
_cookie_transport_allowed
Credentials may cross HTTP only from a local UI to a loopback peer.
backend/api/routers/dub_core.py:50
↓ 1 callers
Function
_copy_atomic
(source: Path, destination: Path)
backend/api/routers/community.py:608
↓ 1 callers
Function
_copy_profile_audio
Copy an imported clip without exposing a partial profile audio file.
backend/api/routers/gallery.py:391
↓ 1 callers
Function
_correct_ordinal_suffix
(n: int)
backend/services/text_normalization.py:397
↓ 1 callers
Function
_count_cudnn8_libs
Count cuDNN 8 shared libraries in the given directory.
scripts/setup.py:196
↓ 1 callers
Function
_crashed_last_session
()
backend/api/routers/system.py:775
↓ 1 callers
Function
_credential_lock
Serialise first-start/renewal across threads and app processes.
backend/worker/tls.py:499
↓ 1 callers
Function
_ct_candidates
(device)
backend/engines/_asr_sidecar/main.py:72
↓ 1 callers
Function
_cuda_arch_supported_for_compile
Check the GPU's architecture against this torch build's arch list. New GPU architectures (e.g. Blackwell sm_120, issue #278) routinely break
backend/services/engine_env.py:153
↓ 1 callers
Function
_cuda_vram_peak_gb
CUDA only. MPS is unified memory (the free-RAM lines already show it) and exposes no peak counter; CPU has no VRAM. Returning None keeps the r
scripts/bench_pipeline.py:131
↓ 1 callers
Function
_cuda_vram_tracking_start
()
scripts/bench_pipeline.py:119
↓ 1 callers
Function
_cudnn8_lib_dir
Return the cuDNN lib subdirectory within the compat install.
scripts/setup.py:189
↓ 1 callers
Function
_current_client_id
The X-OmniVoice-Client-Id of the calling MCP client, if any. FastMCP exposes the HTTP request via its request context on the Streamab
backend/mcp_server.py:157
↓ 1 callers
Function
_data_dir
()
backend/worker/service.py:107
↓ 1 callers
Function
_db_files
Return the set of DB file paths currently present under data_dir.
tests/probe/_boot_runner.py:29
↓ 1 callers
Function
_decode
Turn a settled task into the local branch's return value, or fail.
backend/services/gpu_gateway.py:680
↓ 1 callers
Method
_decode
(self, sample, sample_rate)
omnivoice/scripts/extract_audio_tokens_add_noise.py:235
↓ 1 callers
Method
_decode_and_post_process
Args: tokens: Audio tokens — either a single tensor of shape (num_codebooks, seq_len) or a list of chunk tensors.
omnivoice/models/omnivoice.py:904
↓ 1 callers
Function
_decode_error
(exc: error.HTTPError)
backend/speech_client/__main__.py:37
↓ 1 callers
Function
_decode_minisign_pubkey
Return ``(algorithm, key_id, raw_key)`` from a Tauri-style pubkey. Tauri stores the base64 of the whole two-line minisign ``.pub`` *file*, so
backend/services/gallery.py:197
↓ 1 callers
Method
_decode_online_oneshot
One-shot decode of a whole buffer through the streaming recognizer (for the non-streaming ``transcribe()`` / partial re-decode path).
backend/services/asr_backend.py:1806
↓ 1 callers
Function
_decode_wav
Decode back through ffmpeg so detection sees exactly the published bytes.
scripts/render_gallery.py:83
↓ 1 callers
Function
_deep_import_reason
User-facing reason for a load-time import failure: names the missing module and the repair command (the ``install_hint`` contract of #1185).
backend/services/asr_backend.py:2359
↓ 1 callers
Function
_deep_merge
Tauri-style merge: dicts merge recursively, everything else (incl. lists) is replaced by the platform override.
tests/probe/desktop.py:25
↓ 1 callers
Function
_default_deadline
Worst-case wall time for one attempt, from the shared deadline policy. Same budget the assignment itself carries, so the awaiting side cannot giv
backend/services/gpu_gateway.py:935
↓ 1 callers
Function
_default_dir
()
tests/probe/report.py:242
↓ 1 callers
Function
_default_embedder
Return a real embedder if one is importable, else None (→ skip).
tests/probe/judges/speaker.py:68
↓ 1 callers
Function
_default_timeout
()
backend/services/llm_skills.py:207
↓ 1 callers
Function
_default_uv_cache_root
Volume anchor of uv's default cache (mirrors uv's platform defaults). Only the VOLUME matters — the exact subpath never has to match uv's.
backend/services/sidecar_install.py:254
↓ 1 callers
Function
_deferred_startup
Run the heavy phases behind the already-bound socket. A failure here has import-crash semantics: full traceback to stderr (→ backend_err.log,
backend/main.py:791
↓ 1 callers
Function
_defined_names
Top-level names a module binds: classes, functions, assignments, and whatever it re-exports via its own imports.
tests/test_vendored_omnivoice_imports_1417.py:47
↓ 1 callers
Function
_degraded
(text, slot, tgt)
backend/services/speech_rate.py:280
↓ 1 callers
Function
_delete_cookie_export
Best-effort removal of the per-import authentication export.
backend/services/dub_pipeline.py:915
↓ 1 callers
Method
_deliver_out_of_band
Upload the payload, returning ``([ref], b"")`` on success. Falls back to inline delivery — ``([], payload)`` — only when the payload
backend/worker/transport/client.py:1086
↓ 1 callers
Function
_destination_origin
(connection)
backend/core/csrf.py:104
↓ 1 callers
Function
_detect_cpu_model
Human-readable CPU model. platform.processor() is empty on most Linux distros, so read /proc/cpuinfo there; sysctl on macOS.
backend/api/routers/system.py:45
↓ 1 callers
Function
_detect_gpu
(gpu_name, vram_total_gb) — static for the process lifetime. MPS has unified memory, so there's no separate VRAM figure to report; the name a
backend/api/routers/system.py:64
↓ 1 callers
Function
_detect_system
(tool: str)
backend/services/media_tools.py:490
↓ 1 callers
Method
_device
()
backend/services/subprocess_asr.py:65
↓ 1 callers
Function
_diar_speaker_label
``SPEAKER_00`` → ``Speaker 1`` (mirrors assign_speakers_from_diarization).
backend/services/segmentation.py:731
↓ 1 callers
Function
_dictationFromPrefs
Map a `GET/POST /dictation/prefs` response → the store's dictation fields. * Tolerant of partial/garbage payloads so a malformed response can never w
frontend/src/store/prefsSlice.ts:263
↓ 1 callers
Function
_dictation_wavs
The replay clips DictationDemo.jsx posts to /transcribe.
tests/test_demo_assets_exist.py:86
↓ 1 callers
Function
_dir_size_bytes
Best-effort recursive size (bytes already spent by a partial install).
backend/services/sidecar_install.py:300
↓ 1 callers
Function
_discard
(path: str)
backend/worker/executor.py:1082
↓ 1 callers
Method
_discard_fetched_artifact
( self, attempt: Optional[Attempt], artifact: Optional[str] )
backend/worker/transport/server.py:2539
↓ 1 callers
Method
_discard_terminal_partial
(self, task_id: str, attempt_id: str)
backend/worker/transport/server.py:2839
↓ 1 callers
Function
_discard_tmp
()
backend/api/routers/dub_export.py:1226
↓ 1 callers
Function
_disconnect_after
Async is_disconnected that returns False for the first ``n_ok`` polls (letting those chapters render), then True (client gone).
tests/test_audiobook_cancel.py:64
↓ 1 callers
Function
_discover_orphaned_upload_parts
Find resumable generations no live transport can still own.
backend/worker/transport/server.py:305
↓ 1 callers
Method
_discover_orphans
(self)
backend/worker/inbound/artifacts.py:197
↓ 1 callers
Function
_discover_stored_artifacts
Account retained results, including generations surviving a restart.
backend/worker/transport/server.py:266
↓ 1 callers
Function
_disk_free_gb
()
backend/api/routers/system.py:89
↓ 1 callers
Method
_dispatch_loop
Place queued work on eligible workers. A failed hand-off is not a task failure: the stream may have dropped between selection and sen
backend/worker/service.py:295
↓ 1 callers
Function
_distribution
`[{"name": …, "count": n}, …]`, biggest first. Empty on any error.
backend/services/local_stats.py:46
↓ 1 callers
Function
_do
()
tests/test_fdl_segmented_download.py:49
↓ 1 callers
Function
_do_acquire
()
backend/services/media_tools.py:273
↓ 1 callers
Function
_do_clean_audio
(audio, tmp_dir, clean_id)
backend/api/routers/system.py:968
↓ 1 callers
Function
_do_update_ytdlp
()
backend/services/media_tools.py:591
↓ 1 callers
Function
_do_warmup
()
backend/api/routers/setup/wizard.py:575
↓ 1 callers
Method
_download_artifact_for_session
(self, session, request, context)
backend/worker/transport/server.py:3396
↓ 1 callers
Function
_download_max_workers
Parallel-FILES worker count for snapshot_download (FDL-02). Default 8 — don't crank it: Xet already parallelises *within* each file via concurrent
backend/api/routers/setup/download.py:82
↓ 1 callers
Function
_download_voice_audio
Stream one allow-listed voice clip into an atomic, size-bounded file.
backend/api/routers/community.py:509
↓ 1 callers
Function
_downloaded
Positive absence only: uncertainty and non-HF installs proceed.
backend/worker/capabilities.py:146
↓ 1 callers
Function
_drain
()
tests/test_dub_error_transparency.py:225
↓ 1 callers
Method
_drain
(self, client: WorkerClient)
tests/test_worker_upload_client.py:287
↓ 1 callers
Method
_drain
(self, client: WorkerClient)
tests/test_worker_client_liveness.py:89
↓ 1 callers
Function
_drained
Wait until *count* frames of *kind* have come off the outbox.
tests/test_worker_upload_client.py:314
↓ 1 callers
Function
_drive
(plan, monkeypatch, outputs_dir, *, is_disconnected=None, **kw)
tests/test_audiobook_cancel.py:46
↓ 1 callers
Function
_drive
()
tests/test_gpu_gateway.py:631
↓ 1 callers
Function
_drop_invalid_path_keys
()
backend/core/user_env.py:124
↓ 1 callers
Function
_dur
(pair)
backend/services/speaker_clone.py:341
↓ 1 callers
Function
_durable_params
Parameters safe to persist after inputs have been staged. The live task keeps original paths for a possible local fallback, but the durable r
backend/worker/task_store.py:433
← previous
next →
3,101–3,200 of 15,058, ranked by callers