| | |
|---|
| | | cp "$SHERPA_DLL_DIR/onnxruntime.dll" "$OUT_DIR/" |
|---|
| | | cp "$SHERPA_DLL_DIR/sherpa-onnx-c-api.dll" "$OUT_DIR/" |
|---|
| | | cp "$SHERPA_DLL_DIR/sherpa-onnx-cxx-api.dll" "$OUT_DIR/" |
|---|
| | | cp "$ROOT_DIR/scripts/QA-WINDOWS-PREVIEW.ps1" "$OUT_DIR/" |
|---|
| | | |
|---|
| | | cat >"$OUT_DIR/README-WINDOWS-PREVIEW.txt" <<EOF |
|---|
| | | PrivateVoice Dictation Windows Preview |
|---|
| | |
|---|
| | | How to run: |
|---|
| | | 1. Extract the whole folder. |
|---|
| | | 2. Keep the EXE and DLL files in the same directory. |
|---|
| | | 3. Run "PrivateVoice Dictation.exe". |
|---|
| | | 4. Download/install the SenseVoice model if prompted. |
|---|
| | | 5. Use the configured hotkey and verify text input in Notepad or a browser text field. |
|---|
| | | 3. Right-click "QA-WINDOWS-PREVIEW.ps1" and choose "Run with PowerShell". |
|---|
| | | 4. If PowerShell blocks scripts, run this command from the extracted folder: |
|---|
| | | powershell -ExecutionPolicy Bypass -File .\\QA-WINDOWS-PREVIEW.ps1 |
|---|
| | | 5. Follow the on-screen checklist. It will launch the app, ask you to dictate once, and collect logs. |
|---|
| | | 6. You can also run "PrivateVoice Dictation.exe" directly after QA. |
|---|
| | | |
|---|
| | | Known limitations: |
|---|
| | | - This is not a signed Windows installer. |
|---|
| | | - Real Windows audio, hotkey, paste, tray, and WebView2 behavior still need QA on Windows 10/11. |
|---|
| | | - If Windows Defender or SmartScreen blocks it, allow it only for local testing. |
|---|
| | | - The QA evidence zip may contain app logs with recognized text. Review before sharing if needed. |
|---|
| | | EOF |
|---|
| | | |
|---|
| | | (cd "$OUT_ROOT" && zip -qr "$(basename "$ZIP_PATH")" "$(basename "$OUT_DIR")") |
|---|
