iamdoubz/WhispAssist
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 13 Wiki Activity

13 Releases 13 Tags

RSS Feed
  • v0.7.5 ad3e839b6c
    Compare

    v0.7.5 - Add more models to choose from, updated translations Stable

    iamdoubz released this 2026-07-17 09:15:10 -05:00 | 0 commits to main since this release

    WhispAssist v0.7.5

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    This release is about choice and visibility: pick how strongly your Whisper models are
    quantized (including two new Large v3 Turbo options), see at a glance which meetings are stored
    unencrypted, and convert them to encrypted storage with one click once your vault is unlocked.
    The full interface is also newly translated end-to-end in all 20 languages. Nothing here changes
    the privacy posture; model downloads still come only from the official whisper.cpp repository
    you already use.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    🎛️ Choose your model quantization (Q5 recommended, Q8_0 available)

    Every Whisper model row in Settings ▸ Hardware now has a Quantization dropdown instead
    of a fixed variant:

    • Q5 (Q5_1, or Q5_0 where that's what upstream ships) — the recommended default:
      smaller, faster, near-identical quality.
    • Q8_0 — less aggressively quantized: noticeably larger downloads, slightly higher
      fidelity, for those who want maximum accuracy from a given model size.

    Both variants of a size can be installed side by side; the row's install/activate/remove
    actions apply to whichever variant you've selected.

    🚀 New: Large v3 Turbo models

    Two new catalog entries — large-v3-turbo-q5_0 (~547 MB) and large-v3-turbo-q8_0
    (~833 MB). Large v3 Turbo is OpenAI's pruned large model: the most accurate option in the
    catalog while decoding markedly faster than classic Large. Multilingual only (upstream ships no
    English-only turbo variant), so it also unlocks the language picker.

    🔓 See — and fix — unencrypted meetings

    If you use the at-rest encryption vault, meetings that still have plaintext files on disk
    (recorded before you enabled the vault, or while it was locked) now show an open padlock in
    the meetings list. With the vault unlocked, an Encrypt action appears right on the row and
    seals that meeting's files in place — audio, transcript, notes, summary, everything. Files are
    converted atomically (a crash mid-conversion can't corrupt a meeting), and meetings currently
    recording are protected from conversion. No vault set up? The list stays exactly as before.

    🌍 Complete translations in all 20 languages

    An audit found 53 interface strings (hardware stress test, media-import progress, microphone
    mute, note tools, autostart/tray settings, and more) that had silently fallen back to English
    in every non-English language. All are now properly translated across the 19 non-English
    locales, and an automated parity check keeps every future string translated before it ships.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — the Import dialog links to both.

    1. Download WhispAssist_0.7.5_x64_en-US.msi (or the NSIS WhispAssist_0.7.5_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, does not add itself to startup unless you opt in, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    Deploying to many machines? See docs/enterprise-deployment.md.

    🔐 Checksums (SHA-256)

    8950b7b1ea81a3878975a2e34e7df5733e3d08da2bf92277be580b4c1a61c7d8  WhispAssist_0.7.5_x64_en-US.msi
1d223636c9991a2348de8162e75eb6406fdd1cec16905bd85e39c6effdc19107  WhispAssist_0.7.5_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.7.5_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress; the deployment file never carries secrets. The
    reachable-host allowlist is derived from your settings and enforced in the core.

    Downloads
  • v0.7.4 8cdd0f61df
    Compare

    v0.7.4 - Base CPU model speed increases (92x) Stable

    iamdoubz released this 2026-07-16 10:28:56 -05:00 | 75 commits to main since this release

    WhispAssist v0.7.4

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    A small, focused performance release for CPU transcription. If WhispAssist runs the Whisper
    engine on your CPU — either because your machine has no usable GPU/NPU, or because you selected
    the CPU backend — this update takes it from far-slower-than-real-time to comfortably real-time.
    Machines on the NPU or GPU backends are unaffected in behavior and simply pick up a
    better-optimized build. Nothing here changes the privacy posture; there are no new downloads.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ⚡ CPU transcription fixed (up to ~90× faster on hybrid Intel CPUs)

    Whisper-on-CPU previously spawned one worker thread per logical CPU. On modern hybrid Intel
    processors (P-cores + E-cores + low-power E-cores) the workers synchronize in lock-step, so the
    two slow low-power cores stalled every step while the remaining threads burned CPU waiting —
    transcription ran ~20× slower than real time while looking like it was using the whole machine.

    WhispAssist now caps Whisper's CPU decode at 4 threads (the whisper.cpp upstream default).
    Measured on a Core Ultra 5 135U over a 4-minute test corpus, same accuracy:

    before after
    Speed vs real time 20× slower 4.6× faster (RTF 0.218)
    CPU cost per second of audio 232.6 cpu-sec 0.83 cpu-sec

    Live (streaming) transcription shares this code path and benefits equally. If you want to
    experiment, the WA_WHISPER_THREADS environment variable overrides the cap.

    🔧 Native code now built fully optimized

    A build-toolchain issue caused the bundled native libraries (including whisper.cpp's CPU
    kernels) to be compiled without compiler optimization. That's fixed; this release ships
    optimized native code across the board. (For CPU transcription the thread fix above is the
    dominant win; this one is belt-and-suspenders for everything else.)

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — the Import dialog links to both.

    1. Download WhispAssist_0.7.4_x64_en-US.msi (or the NSIS WhispAssist_0.7.4_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, does not add itself to startup unless you opt in, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    Deploying to many machines? See docs/enterprise-deployment.md.

    🔐 Checksums (SHA-256)

    5c4e6d6b23a4b404c6fa200b6491e74f143c3a886042dc3bef3c2861168ef17c  WhispAssist_0.7.4_x64_en-US.msi
eb2ececff06373e9cbb43ee6455c433c1b88785e9060b68ef7bed193191d9761  WhispAssist_0.7.4_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.7.4_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress; the deployment file never carries secrets. The
    reachable-host allowlist is derived from your settings and enforced in the core.

    Downloads
  • v0.7.3 87fb665e9e
    Compare

    v0.7.3 - Intel NPU/iGPU speed improvements ~4x Stable

    iamdoubz released this 2026-07-15 20:46:53 -05:00 | 99 commits to main since this release

    WhispAssist v0.7.3

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    A performance release: on Intel Core Ultra machines, transcription now leaves the CPU nearly
    idle    . The Whisper encoder keeps running on the NPU; the decoder — previously the dominant
    CPU cost — now decodes with a KV cache and runs on the Intel iGPU. Nothing here changes
    the privacy posture: the one new download is the upgraded transcription model, fetched from the
    same source as before.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ⚡ Faster, cooler transcription on NPU systems

    KV-cache decoding

    The NPU engine previously re-processed the entire token sequence for every new token it
    generated. It now uses the upgraded merged Whisper decoder and caches attention state between
    steps, so each step does a small, constant amount of work. This benefits every machine using
    the NPU/DirectML engine, whatever device the decoder runs on.

    Decoder on the Intel iGPU

    On NPU systems the decoder now runs on the Intel iGPU via the same OpenVINO runtime already
    installed for the NPU — no new runtime download. Division of labor while transcribing:

    • Encoder → NPU (the short periodic spikes on Task Manager's NPU graph)
    • Decoder → Intel iGPU
    • CPU → audio plumbing only

    Measured on a Core Ultra 5 135U (base.en): decoding on the iGPU ran ~4.7× faster than the
    same KV-cache decode on the CPU (1.29 s vs 6.10 s for a 19 s clip), with identical transcripts —
    and that work no longer occupies the CPU at all.

    Fallbacks: no usable Intel iGPU → the decoder automatically drops back to the CPU (still
    with the KV-cache speedup), and the log says so. Machines with an NVIDIA/AMD GPU can keep using
    the Vulkan backend (Settings ▸ Hardware), which runs the whole model on that GPU. The startup
    log names both devices: transcription engine: ONNX base.en (encoder EP: OpenVINO/NPU, decoder EP: OpenVINO/GPU).

    Model update (automatic, one-time)

    The engine's decoder artifact changed to the merged export (decoder_model_merged.onnx,
    ~208 MB, same trusted source). WhispAssist fetches it automatically in the background on first
    launch after updating; until it arrives, transcription simply uses the whisper.cpp engine, as on
    a fresh install. The superseded decoder file is cleaned up automatically.

    ✨ Small addition

    Settings ▸ Calendar now has a one-click libpst (readpst) download button — the same style
    as the Import dialog's ffmpeg/yt-dlp buttons — linking to the ezwinports Windows build.
    .pst calendar import shells out to readpst, which isn't bundled; the button just opens the
    download page in your browser.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — the Import dialog links to both.

    1. Download WhispAssist_0.7.3_x64_en-US.msi (or the NSIS WhispAssist_0.7.3_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, does not add itself to startup unless you opt in, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    Deploying to many machines? See docs/enterprise-deployment.md.

    🔐 Checksums (SHA-256)

    91a32a17addc31fbc00f38966d52402ca0c5e9e7a581a0431c383ca5a60532bd  WhispAssist_0.7.3_x64_en-US.msi
764b758544b075b5aefe9dde37b04e4966df528912fb884ed1be5bd57905ecdb  WhispAssist_0.7.3_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.7.3_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress; the deployment file never carries secrets. The
    reachable-host allowlist is derived from your settings and enforced in the core.

    Downloads
  • v0.7.2 21f3d76ed0
    Compare

    v0.7.2 - Import meeting, better notes, mute mic, close to system tray and more! Stable

    iamdoubz released this 2026-07-15 18:16:16 -05:00 | 110 commits to main since this release

    WhispAssist v0.7.2

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    A bug-fix release focused on speaker identification. The live transcript now labels you
    correctly from the very first diarization pass, your words no longer get filed under other
    speakers, renaming a speaker finally updates your Notes (not just the transcript), and action
    items are readable at any pane width. Nothing here changes the privacy posture: everything optional
    stays off-by-default and local-first.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ✨ New

    Add a meeting — now in the background, with a progress tracker

    Importing a recording (a local audio/video file or a YouTube/streaming/direct URL) no longer blocks
    the app while it works. Click Import and the meeting appears in your list immediately with a
    four-step progress tracker — Transcode → Transcribe → Identify speakers → Finalize — where the
    current step pulses and finished steps show how long they took. A 25-minute video that used to lock
    the window for ~13 minutes now transcribes quietly in the background.

    • Pick the transcription model right in the dialog, and see "Transcribed with …" on the
      finished meeting so you always know how it was produced.
    • One-click links to download ffmpeg and yt-dlp (still external, not bundled).
    • A failed import stays in your list marked error instead of vanishing.

    Better notes

    • AI-enhance (✨): turn rough notes into clean, structured notes using your local LLM, with a
      one-step Undo. Off unless you have a local model configured.
    • Slash commands & a formatting toolbar: type /todo, /h1, /quote, … or use the toolbar for
      headings, lists, checkboxes, quotes, and dividers.
    • Fix: notes no longer show stale text after re-transcribing a meeting — the pane updates in
      place, no restart needed.

    Mute your microphone — press M

    Mute/unmute the mic mid-meeting with the M key or the new mic button by the level meter. The mic
    channel goes silent (recording, live transcript, and meter) while system/loopback audio keeps
    capturing.

    Close to system tray

    Closing the window now keeps WhispAssist running in the background instead of quitting. Reopen it
    from the tray icon; the tray's Quit exits fully. On by default — toggle it in
    Settings ▸ Recording ▸ Close to system tray. (This release also fixes a bug that showed two
    WhispAssist icons in the tray — there's now just one.)

    Privacy & hardware odds and ends

    • Vault lock card in Settings ▸ Privacy: lock/unlock the encrypted store and change its
      password at a glance.
    • Test your audio devices: a live level meter for your mic and system audio, plus a test tone.
    • Quick hardware stress test: benchmark the available backends against your installed models and
      apply the fastest real-time combination.
    • Tidier recording header.

    🐛 Fixed

    "You" now shows up live, not just after the meeting

    With the microphone enabled, the live transcript used to show every line as a numbered speaker
    until you pressed Stop. The live pass now uses the same channel-based attribution as the final
    pass — your mic is the "You" channel — so your lines are labeled You from the first
    15-second diarization tick. (Reminder: with the microphone disabled in Settings there is no mic
    channel, so "You" can't appear at all.)

    Your words no longer end up under other speakers

    Speaker attribution used to hand each transcript segment to the single longest overlapping
    diarization span, so a long far-side span could swallow a segment you spoke most of. Segments are
    now attributed by comparing total voiced time per audio channel (your mic vs. the meeting
    audio), and ties go to You — your voice is physically separated on its own channel, so channel
    evidence wins. Renames you make during a recording also survive to the finalized meeting now
    instead of being overwritten by the automatic labels.

    Renaming a speaker updates the Notes pane

    Speaker names are written into notes.md when a meeting finalizes — and renaming a speaker
    afterwards (typed name or attendee pick) never touched them. Renames now rewrite the dialogue tags
    in notes.md too, your own note edits are preserved, and the open Notes editor refreshes in place.

    Action items you can actually read

    After generating a summary, action-item text was crushed to near-zero width by the six controls
    sharing one row in the right pane. Each item is now a two-row card: the action text gets the full
    first row, with owner, due date, and reminder on a second row beneath it — readable without
    resizing anything.

    Honest engine reporting (NPU)

    Meetings transcribed on the Intel NPU claimed the configured whisper model (e.g.
    medium.en-q5_0 · npu) when the NPU engine actually runs its own ONNX model. The meeting header
    now reports the model truly used — base.en (onnx) · npu — and the engine logs which execution
    provider the encoder is running on at startup.

    Why the NPU graph reads 0% in Task Manager: only Whisper's encoder runs on the NPU, as
    short bursts (the periodic spikes on the NPU graph); the decoder runs on the CPU by design.
    Low average NPU utilization while transcribing is expected and does not mean the NPU is unused —
    the meeting header's · npu badge and the startup log are the source of truth.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — the Import dialog links to both.

    1. Download WhispAssist_0.7.2_x64_en-US.msi (or the NSIS WhispAssist_0.7.2_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, does not add itself to startup unless you opt in, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    Deploying to many machines? See docs/enterprise-deployment.md.

    🔐 Checksums (SHA-256)

    7c94854341e42dff73988d9cc3ff15c6f30aa02fb9b478de9fb4c63197f2f1dc  WhispAssist_0.7.2_x64_en-US.msi
a0adc2a61e6e1197469acacb78fdce96176edb22cb6125912cb982a3b7739002  WhispAssist_0.7.2_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.7.2_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress; the deployment file never carries secrets. The
    reachable-host allowlist is derived from your settings and enforced in the core.

    Downloads
  • v0.6.0 2c8958da01
    Compare

    v0.6.0 - Vastly improved microphone and speaker voice parsing Stable

    iamdoubz released this 2026-07-14 15:52:52 -05:00 | 171 commits to main since this release

    WhispAssist v0.6.0

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    A big release for speaker labeling. WhispAssist now records your microphone and the computer's
    audio on separate channels, so it can tell "You" apart from each "Speaker" reliably —
    and that labeling now survives re-transcription. Along the way this release fixes a cluster of
    recording and playback bugs. Everything optional stays off by default and local-first; no
    privacy changes.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ✨ Highlights

    Reliable "You" vs "Speaker" labeling (dual-channel recording)

    When your microphone is on, WhispAssist now records the meeting with your mic on one channel and
    the computer/system audio on the other    , keeping the two voices separate in the file. Diarization
    runs on the system-audio side alone to split the other participants, while your own speech is
    labeled "You" directly     — it's never mixed in with the others. Because the separation lives in the
    recording itself, re-transcribing a meeting keeps the correct "You" / "Speaker N" labels instead
    of guessing again. (Playback folds the two channels together so it sounds normal; exported copies
    stay ordinary stereo.)

    Speaker labeling fixes

    • No more speaker explosion — a two-person conversation no longer gets split into dozens of
      "speakers."
    • Re-transcribe keeps distinct speakers — re-running a transcript with a bigger model no longer
      collapses everyone into a single speaker.
    • The Participants list matches the transcript — after re-transcribing, it no longer shows a
      stale, inflated speaker count.
    • Live "You" during recording — the mic speaker is identified as "You" while you record, not
      only after you stop.

    🐛 Recording & playback fixes

    • Your voice is captured during quiet stretches — previously, when nothing was playing through
      your speakers, your own speech could be dropped from the recording. It's now recorded throughout.
    • Playback glitches fixed — recordings no longer develop clicks/ticks or a "robotic," fast-
      forward-like stretch late in a meeting (a microphone-buffering bug in the recorder).

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — install those separately.

    1. Download WhispAssist_0.6.0_x64_en-US.msi (or the NSIS WhispAssist_0.6.0_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, never adds itself to startup, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    Note on existing recordings: the improved labeling applies to meetings recorded with v0.6.0.
    Recordings made with earlier versions still re-transcribe using the previous best-effort method.

    🔐 Checksums (SHA-256)

    4974F6D234F886479AC6AC76D311163EF4E62A3AA6280DC1AD549F51060C30CF WhispAssist_0.6.0_x64_en-US.msi
291FFCBDEE917FD9DE101449B2A00C10ABB54A89D3184C83E59035CAB0CB5DD1 WhispAssist_0.6.0_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.6.0_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress. The reachable-host allowlist is derived from your
    settings and enforced in the core.

    Downloads
  • v0.5.2 4f3af98bfa
    Compare

    v0.5.2 - Bug fix (broken migrations) Stable

    iamdoubz released this 2026-07-13 16:18:10 -05:00 | 216 commits to main since this release

    WhispAssist v0.5.2

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    A small but important bug-fix release. v0.5.2 fixes a startup crash that could stop WhispAssist
    from opening after an upgrade, and makes any future database problem show a clear message instead
    of failing silently. No feature or privacy changes — everything optional stays off-by-default and
    local-first.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    🐛 Fixes

    Fixed: app failing to launch after an upgrade

    On some machines, upgrading could leave the app unable to open at all — no window, no message,
    just nothing. The cause was a database-migration mismatch during startup: WhispAssist would hit
    the error while opening its local database (wa.db) and, because the failure happened before the
    window existed, the process exited silently and Windows reported only a generic crash.

    Startup database errors now explain themselves

    Instead of that silent exit, a failure to open wa.db now shows a native error dialog naming
    the problem and pointing at the database location, then exits cleanly. Your recordings and notes
    are never touched — the message tells you exactly what happened and how to recover, rather than
    leaving you with an app that won't start.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — install those separately.

    1. Download WhispAssist_0.5.2_x64_en-US.msi (or the NSIS WhispAssist_0.5.2_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, never adds itself to startup, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    🔐 Checksums (SHA-256)

    b173b6b15590a65e96d3b24d0ad97666b8d1da905169d807ecc1c51df0584d3e  WhispAssist_0.5.2_x64_en-US.msi
441b2e7ec7c68be39f9d2381af59ab9eadf916d6d23d0c5541d78bdd88c588b2  WhispAssist_0.5.2_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.5.2_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress. The reachable-host allowlist is derived from your
    settings and enforced in the core.

    Downloads
  • v0.5.1 226593b49d
    Compare

    v0.5.1 - Translations available in 20 additional flavors! Stable

    iamdoubz released this 2026-07-13 08:49:36 -05:00 | 223 commits to main since this release

    WhispAssist v0.5.1

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    v0.5.0 put the i18n groundwork in place. v0.5.1 fills it in: WhispAssist now ships in
    20 languages, so you can run the whole app — the shell, meetings list, transcript/notes, summary
    panel, and every corner of Settings — in your own language. Everything stays off-by-default and
    local-first; this is a UI-language release with no change to what leaves your device (nothing, by default).

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ✨ Highlights

    20 interface languages

    Pick your language under Settings ▸ Language. Alongside English, this release adds full
    translations for:

    • Arabic (العربية) · Bengali (বাংলা) · German (Deutsch)
    • Spanish — Spain (Español, España) and Mexico (Español, México)
    • Finnish (Suomi) · French — France (Français, France) and Canada (Français, Canada)
    • Hindi (हिन्दी) · Korean (한국어) · Norwegian Bokmål (Norsk bokmål)
    • Portuguese — Brazil (Português, Brasil) and Portugal (Português, Portugal)
    • Russian (Русский) · Sinhala (සිංහල) · Swedish (Svenska)
    • Tamil (தமிழ்) · Urdu (اردو) · Mandarin Chinese, Simplified (中文简体)

    Your choice persists across launches, and any untranslated string quietly falls back to English rather
    than showing a raw key — so partial translations degrade gracefully.

    Right-to-left layout

    Selecting Arabic or Urdu flips the whole interface to right-to-left, so those languages
    read and lay out correctly rather than being crammed into an LTR shell.

    Adding a further language remains a one-file job — drop in a single JSON translation, no code changes.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — install those separately.

    1. Download WhispAssist_0.5.1_x64_en-US.msi (or the NSIS WhispAssist_0.5.1_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, never adds itself to startup, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    🔐 Checksums (SHA-256)

    911618e6996a079bad1cbe46731c89ee247f3f5f17c09b60091d7e6a5e7cbef0  WhispAssist_0.5.1_x64_en-US.msi
ffabbd74def981da176931349b2813a3927d7376c0d18e9d191891abfc17c385  WhispAssist_0.5.1_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.5.1_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress. The reachable-host allowlist is derived from your
    settings and enforced in the core.

    Downloads
  • v0.5.0 5f300ee177
    Compare

    v0.5.0 - Export and import, i18n, Obsidian, and more! Stable

    iamdoubz released this 2026-07-12 14:30:32 -05:00 | 249 commits to main since this release

    WhispAssist v0.5.0

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    This release is about living with your recordings: play back a meeting while following along in
    the transcript, manage action items by hand, move recordings between computers, drop a meeting into
    your Obsidian vault, and — new in this release — use WhispAssist in your own language. Everything
    stays off-by-default and local-first.

    One universal installer (MSI and NSIS) covers every machine: Vulkan for all GPUs
    (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ✨ Highlights

    Click the transcript to play that moment

    The recording player and the transcript now talk to each other. Click any transcript line to jump
    the audio to that moment     and start playing, and as playback runs the current line highlights and
    scrolls into view     so you never lose your place. If you scroll by hand, auto-scroll steps aside for
    a few seconds so it doesn't fight you.

    Interface language selector (i18n)

    WhispAssist can now be fully translated. A new Settings ▸ Language picker switches the
    interface language, English ships as the baseline, and every user-facing string across the app —
    the shell, meetings list, transcript/notes, summary panel, and all of Settings — now flows through a
    single translation layer. Adding a new language is as simple as translating one JSON file; no
    code changes. (This release ships English; the groundwork is in place for community translations.)

    Manage action items yourself

    Action items are no longer just whatever the summary extracted. You can now add, edit, and delete
    them by hand     in the summary panel, set an owner and due date, and toggle a local reminder. Your
    edits are the source of truth and are reconciled cleanly — deleting an item also cancels its reminder.

    Move recordings between computers (Export / Import)

    A new Export & import section in Settings ▸ Storage writes each meeting as a portable
    bundle folder (audio, transcript, notes, summary, and a meeting.json manifest) and imports them
    back on another machine. Imported meetings get a fresh id, so re-importing never overwrites anything.
    Export to any folder — a synced drive, a USB stick, or a sync target's local mount — and carry it across.

    Export a meeting to Obsidian

    A new Obsidian export writes a single self-contained vault note — YAML frontmatter
    (title, date, duration, participants, tags) plus notes, summary, action items, and a timestamped
    transcript — without the audio. Drop it in your vault and the transcript is fully searchable.

    🚀 Also new since v0.4.0

    • Per-segment transcript timestamps — every line now shows a quiet m:ss (or h:mm:ss) time
      prefix, in both the live and finalized views.
    • Auto-resync on edit — when sync is enabled, editing a meeting's notes, summary, transcript,
      tags, or action items re-uploads just the changed artifacts automatically (deduped by hash, so
      an unchanged save uploads nothing). Off unless sync is configured.

    🐛 Fixes & polish

    • Transcript scroll-intent handling refined so playback auto-scroll never yanks you back while you're
      reading.
    • Import preserves each meeting's original date rather than stamping the import time.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — install those separately.

    1. Download WhispAssist_0.5.0_x64_en-US.msi (or the NSIS WhispAssist_0.5.0_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, never adds itself to startup, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    🔐 Checksums (SHA-256)

    b423feed1a1171384e46c5e0b5aa63cbf912bfe8f29cc89a5a027bb15358075c  WhispAssist_0.5.0_x64_en-US.msi
2dce26b5603a5d02094533c594d814a3d72e05512e1d735ea69e1e309f15b710  WhispAssist_0.5.0_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.5.0_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress. The reachable-host allowlist is derived from your
    settings and enforced in the core.

    Downloads
  • v0.4.0 57cd46be46
    Compare

    v0.4.0 - Notes pane, import meeting from URL, and much much more! Stable

    iamdoubz released this 2026-07-11 09:28:52 -05:00 | 318 commits to main since this release

    WhispAssist v0.4.0

    Privacy-first, Windows-native meeting assistant — everything on-device, nothing leaves unless you say so.

    This release is a big step up from v0.2.0. It makes the live transcript feel natural, lets you
    import existing recordings, adds hands-free recording tied to your calendar, ships the local
    MCP server and feature briefs for handing meetings to your coding agents, adds hosted-AI
    providers     and multilingual transcription, and rounds out Outlook calendar support — all
    still off-by-default and local-first.

    One universal signed installer (MSI and NSIS) covers every machine: Vulkan for all
    GPUs (NVIDIA/AMD/Intel), the Intel NPU (OpenVINO), a DirectML fallback, and CPU.

    ✨ Highlights

    Smoother, more natural live transcript

    The live transcript no longer appears in rigid ~4-second blocks. It now streams a growing line
    that refreshes about once a second and commits at natural pauses    , so words show up as they're
    spoken and sentences are no longer chopped across lines. The in-progress line is shown lightly
    until it settles. The Low overhead preset uses a gentler cadence to save CPU.

    Add a meeting from a file or URL

    A new “Add meeting” button imports an existing recording — a local audio/video file or a
    URL (YouTube, a streaming page, or a direct media link). WhispAssist transcribes and diarizes
    it exactly like a live meeting. Uses ffmpeg (and yt-dlp for URLs), which you install
    yourself — neither is bundled.

    Auto-start recording on calendar events

    Opt-in: when an imported calendar event begins while WhispAssist is open, recording can start
    automatically     (using your default retention setting). It’s a single one-shot timer — nothing
    polls in the background    , keeping idle resource use at zero.

    Live dual level meter (system + microphone)

    While recording, the level meter now overlays the system and microphone signals in two colours,
    so you can see at a glance that both sides of the conversation are being captured.

    Speaker diarization is now one click to enable

    The offline diarization engine shipped earlier, but the models weren’t installable from the UI.
    Settings ▸ Hardware now has download cards for the segmentation + embedding models. Install
    them and finished recordings are split by speaker (Speaker 1, Speaker 2, …); your own mic voice is
    auto-labelled from a short voiceprint.

    🚀 Also new since v0.2.0

    • Local MCP server + feature briefs — hand meeting context to your own coding agents (Claude,
      Codex, Copilot, OpenCode) over a loopback-only, token-gated MCP server that is off by
      default, scope-limited, audited, and adds no egress of its own. Distil a meeting into an
      agent-ready feature brief. (Note: Granola recently added an MCP server too, but it’s
      cloud-hosted; WhispAssist’s is local.)
    • Optional hosted AI providers — Anthropic (Claude) and OpenAI-compatible endpoints behind the
      same LlmProvider interface, off by default, with a one-time “this leaves your device” notice;
      keys live in the OS credential store.
    • Multilingual transcription — choose a transcription language (or auto-detect) in
      Settings ▸ Hardware with multilingual Whisper models; each meeting records the language it
      actually used, and you can re-transcribe with a different model/language.
    • Outlook .pst / calendar — import events and attendees from a local .pst (read-only,
      range-limited, de-duplicated, with cleanup), attach meetings to events, plus Microsoft Graph
      calendar via OAuth.
    • Granola-style live notes — a typable Notes pane during recording, click a transcript line to
      attach a note to that moment, resizable/hideable panes.
    • Sync additions — Dropbox and Box targets (OAuth), and chunked/resumable upload for large
      recordings, alongside the existing WebDAV set and client-side encrypt-before-upload.

    🐛 Fixes & polish

    • Drop whisper-hallucinated filler on silent windows (“you”, “Thank you.”).
    • readpst now runs headless (no console window popping up on import).
    • Stable, de-duplicated calendar event keys; collapse pre-existing duplicate rows.
    • Full-text search reindexes on tags/summary changes.
    • Case-insensitive language matching; corrected medium-model quantization id.

    📦 Install

    Requirements: Windows 10 or 11 (x64). WhispAssist needs WebView2 (preinstalled on Windows 11;
    the installer fetches it on Windows 10). Importing from a file/URL additionally needs ffmpeg
    (and yt-dlp for URLs) on your PATH — install those separately.

    1. Download WhispAssist_0.4.0_x64_en-US.msi (or the NSIS WhispAssist_0.4.0_x64-setup.exe).
    2. Run it and accept the UAC prompt. If SmartScreen appears, choose More info → Run anyway.
    3. Launch WhispAssist from the Start menu.

    On first run WA picks the best transcription backend (NPU → NVIDIA → AMD → Intel → CPU). It runs
    without admin rights, never adds itself to startup, and keeps all data under
    %LOCALAPPDATA%\WhispAssist.

    🔐 Checksums (SHA-256)

    b9b3ed5cf6ba75a3aaa1ae5d9fce816417a3a57a0e357bc94660cc125553853c  WhispAssist_0.4.0_x64_en-US.msi
3602cd07971378aa0b4d9c04b2c87ecf29e45d104f2ac126b20e7b52d5a7df60  WhispAssist_0.4.0_x64-setup.exe

    Verify after download:

    Get-FileHash .\WhispAssist_0.4.0_x64_en-US.msi -Algorithm SHA256

    Privacy, unchanged

    Everything optional is off by default. With nothing configured, WhispAssist makes no content
    egress at all    . Recording is opt-in; sync/AI credentials live only in the OS credential store; the
    MCP server is loopback-only and adds no egress. The reachable-host allowlist is derived from your
    settings and enforced in the core.

    Downloads
  • v0.3.0 9f1fc671da
    Compare

    v0.3.0 - MCP server, audio device selection, multi-language speech recognition Stable

    iamdoubz released this 2026-07-07 22:06:10 -05:00 | 384 commits to main since this release

    What's Changed

    • Feature chore bug 005 in #19
    • Update doc roadmap in #18

    Contributors

    Checksums

    • b7610280983d6506d83ec0e4ea4dcb25bf1c2d73c26c93e6d546730b6d91808f WhispAssist_0.3.0_x64_en-US.msi
    • 45e002b44be87a76239621f5af419b3e85dca8597a3915e45b55254e02e9bf4b WhispAssist_0.3.0_x64-setup.exe

    Full Changelog: v0.2.0...v0.3.0

    Downloads