iamdoubz/WhispAssist
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 13 Wiki Activity
213 Commits 1 Branch 13 Tags
281ccf0b182630b747d429c5d43baad25d34048d
Commit Graph
213 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
iamdoubz 281ccf0b18 feat(hardware): directml_would_help() gate for the Settings package card (P2) 
True only when staging DirectML would light up a present GPU this build can't
otherwise accelerate: npu compiled, no Vulkan baked in, and an AMD/Intel GPU
present (or NVIDIA without CUDA).
 2026-07-06 07:21:50 -05:00
iamdoubz ab5aa4e986 feat(commands): register download_directml_package command (P2) 2026-07-05 19:40:28 -05:00
iamdoubz b69c91a598 feat(commands): route AMD/Intel to DirectML ONNX engine + download_directml_package (P2) 
load_transcriber dispatches OnnxOpenVino/OnnxDirectML through one OnnxTranscriber
block. download_and_extract_runtime is parameterized (sha + readiness) so the new
stage_directml_runtime reuses it; hardware_status reports a directml package field.
 2026-07-05 19:40:27 -05:00
iamdoubz 0f2940e52b feat(hardware): make directml_ready() real (npu feature + staged runtime) (P2) 2026-07-05 19:40:27 -05:00
iamdoubz 9ae02cabfe docs(transcription): update OnnxTranscriber name in doc comments (P2) 2026-07-05 19:40:17 -05:00
iamdoubz eb87bf4edb refactor(transcription): re-export OnnxTranscriber; note ONNX tier covers NPU+DirectML (P2) 2026-07-05 19:40:17 -05:00
iamdoubz 0f93f5947e feat(transcription): generalize OnnxNpuTranscriber -> OnnxTranscriber with DirectML EP (P2) 
load() now selects OpenVINO (NPU) or DirectML (AMD/Intel GPU) by BackendId over
the same ONNX artifacts; ensure_runtime_env takes the target runtime DLL. Adds
an opt-in directml_transcribes_speech spike for on-GPU validation.
 2026-07-05 19:40:16 -05:00
iamdoubz 2e8bb730a3 feat(paths): DirectML runtime dir/dll/ready helpers (P2) 2026-07-05 19:40:16 -05:00
iamdoubz c4658efe28 build(deps): enable ort DirectML EP for the non-Vulkan GPU path (P2) 2026-07-05 19:40:15 -05:00
iamdoubz 7fe0d87e55 feat(transcription): drive engine choice from AccelPath resolver (P0) 
load_transcriber now resolves the accel path instead of assuming a GPU
BackendId has a working offload. Only requests whisper GPU offload for
WhisperCuda/WhisperVulkan; NPU goes to the ONNX engine; everything else decodes
on CPU and reports the backend actually used.
 2026-07-05 19:27:28 -05:00
iamdoubz e9567b823a feat(hardware): AccelPath resolver + honest GPU availability (P0) 
whisper.cpp bakes in one GPU backend at compile time, so a BackendId alone
can't say how a vendor is served. Add AccelPath + resolve_accel(_with) as the
single source of truth over (compiled features x runtime readiness), and derive
BackendInfo.available from it in the DXGI + NPU paths. Closes the gap where a
present-but-unsupported GPU was marked available and best() routed to a no-op
GPU. Pure resolve_accel_with is unit-tested across all vendor/feature combos.
 2026-07-05 19:27:23 -05:00
iamdoubz a239d10a01 Merge pull request 'Feature again here' (#13) from feature_again_here into main 
Reviewed-on: #13
 2026-07-05 18:20:16 -05:00
iamdoubz 95b2635607 Add npu 2026-07-05 18:19:22 -05:00
iamdoubz c24feeb542 docs(readme): reflect working v0.1.4 (CPU/NPU/Vulkan, vault, sync) + refresh Granola/Meetily comparison 2026-07-05 17:58:26 -05:00
iamdoubz fa96a16be4 chore: normalize Cargo.toml line endings after BOM-strip pass 2026-07-05 00:42:39 -05:00
iamdoubz 76c33fd772 fix(build): strip UTF-8 BOM from package.json/tauri.conf.json (broke vite type:module + vitefu JSON.parse) 2026-07-05 00:42:19 -05:00
iamdoubz 68e4d2fece chore(release): bump version to 0.1.4 2026-07-05 00:36:33 -05:00
iamdoubz 02a215c2e3 test(transcription): GPU/CPU timing spike (opt-in) for whisper.cpp backends 2026-07-05 00:36:33 -05:00
iamdoubz 2e03eab8f8 chore(release): bump version to 0.1.3 2026-07-04 09:22:06 -05:00
iamdoubz 8aa6ba59a7 feat(ui): redesign AI status/actions + collapsible advanced Ollama config (T5.2) 2026-07-04 09:21:29 -05:00
iamdoubz a0ce2e2321 feat(ui): Ollama parameter catalog (defaults/ranges/help) as single source of truth 2026-07-04 09:21:28 -05:00
iamdoubz b980c833c4 feat(llm): type llm_advanced on AppSettings 2026-07-04 09:21:28 -05:00
iamdoubz 606bf67943 feat(llm): wire llm_advanced through provider construction + defaults (T5.2) 2026-07-04 09:21:27 -05:00
iamdoubz 34e511bcf6 feat(llm): inject system prompt + think/keep_alive/options into Ollama requests (T5.2) 2026-07-04 09:21:27 -05:00
iamdoubz 0165d4752c feat(llm): add sparse llm_advanced settings for Ollama tuning (T5.2) 2026-07-04 09:21:27 -05:00
iamdoubz 59e457c421 Merge pull request 'Feature cleanup' (#12) from feature_cleanup into main 
Reviewed-on: #12
 2026-07-03 11:42:34 -05:00
iamdoubz 0d335a6880 chore(release): bump version to 0.1.2 2026-07-03 11:40:15 -05:00
iamdoubz 4fc271dd1a feat(ui): AI settings section — pick provider, set LAN/local endpoint + model, test (FR-LLM-1) 2026-07-03 09:49:35 -05:00
iamdoubz df7ea29c67 feat(llm): LLM status + setLlmProvider in settings store 2026-07-03 09:49:35 -05:00
iamdoubz 1f20ec13ab fix(llm): setLlmProvider arg key (was config, command expects args) 2026-07-03 09:49:34 -05:00
iamdoubz 78335b3e69 feat(llm): treat private LAN endpoints (RFC-1918) as local, not internet egress (FR-LLM-6) 2026-07-03 09:49:34 -05:00
iamdoubz 2093b39fe1 docs(npu): runtime bundle README (contents, sha, hosting) 2026-07-03 09:18:12 -05:00
iamdoubz 62283edec2 chore: gitignore the NPU runtime zip (was wrongly under Vite's dist/) 2026-07-03 09:18:11 -05:00
iamdoubz a38bf20e79 feat(npu): download+unzip the hosted OpenVINO runtime bundle w/ progress + sha256 (T3.4) 2026-07-03 09:16:41 -05:00
iamdoubz 7cc437d1b5 build(npu): add zip crate to extract the runtime bundle (T3.4) 2026-07-03 09:16:40 -05:00
iamdoubz 164cecae7a chore: gitignore dist/ (build artifacts) 2026-07-03 09:16:40 -05:00
iamdoubz 368be69ff7 chore(release): bump version to 0.1.1 2026-07-03 09:06:43 -05:00
iamdoubz 4001d38b9b fix(npu): ship npu in the default build so the shipped app can use/download it (T3.4); bump 0.1.1 2026-07-03 09:06:42 -05:00
iamdoubz 6b215cae36 fix(ui): show real error message on NPU download + vault failures 2026-07-03 09:06:42 -05:00
iamdoubz 88e1fcf1d5 fix(ui): show real error message (not [object Object]) on OAuth link failure 2026-07-03 09:06:42 -05:00
iamdoubz 00ac43c705 Merge pull request 'Feature another one' (#11) from feature_another_one into main 
Reviewed-on: #11
 2026-07-03 08:56:28 -05:00
iamdoubz 06c1764506 version 0.1.0 2026-07-03 08:55:16 -05:00
iamdoubz 713ab381d0 chore(release): bump version to 0.1.0 2026-07-03 08:48:59 -05:00
iamdoubz bb03cd2328 feat(ui): encrypt-before-upload checkbox in sync target form (T9.12) 2026-07-03 08:47:36 -05:00
iamdoubz ce562ca67a feat(ui): startup vault-unlock gate (T8.8) 2026-07-03 08:47:36 -05:00
iamdoubz 1824475cef feat(sync): seal recording on finalize; encrypt-before-upload in pump; honor config flag (T9.12/T8.8) 2026-07-03 08:47:36 -05:00
iamdoubz 97bda924b3 feat(vault): decrypt sealed recordings on read (audio at-rest) (T8.8) 2026-07-03 08:47:36 -05:00
iamdoubz 1b78d4db0b feat(vault): ensure_sealed for encrypt-before-upload; zeroize keys on lock (T9.12/T8.8) 2026-07-03 08:47:35 -05:00
iamdoubz 9505935490 build(vault): add zeroize for in-memory key wiping (T8.8 hardening) 2026-07-03 08:47:35 -05:00
iamdoubz 0b7461dc9d feat(ui): encryption-vault panel in Settings privacy (enable/unlock/lock/change) (T8.8) 2026-07-03 08:06:06 -05:00