Skip to content

Releases: BoltzmannEntropy/MimikaStudio

MimikaStudio v2026.04.1

01 Apr 10:39
@BoltzmannEntropy BoltzmannEntropy
v2026.04.1
66c02d5

Choose a tag to compare

View all tags
MimikaStudio v2026.04.1 Latest
Latest

MimikaStudio v2026.04.1 Release Notes

Release Date: April 1, 2026
Platform: macOS (Apple Silicon)

What's New In v2026.04.1

  • Added embedded PDF page preview inside the Audiobooks document pane, with page navigation and zoom controls alongside extracted text preview.
  • Disabled the old 7-day expiration messaging and removed active Polar.sh and LemonSqueezy checkout or portal flows from the in-app Pro screen.
  • Removed website pricing and buying paths so downloads now point directly to GitHub releases without purchase messaging.

User Impact

  • Users can inspect PDF pages directly before generating audiobooks instead of relying only on extracted text.
  • Users are no longer told that access expires after 7 days, and purchase links are clearly disabled in this build.
  • Website visitors now see direct-download messaging instead of pricing cards or checkout prompts.

Technical Notes

  • Extended flutter_app/lib/screens/audiobook_screen.dart to support dual PDF or text preview modes using Syncfusion's PDF viewer controller.
  • Simplified flutter_app/lib/screens/pro_screen.dart so it only reports current license state and accepts manual activation of existing keys.
  • Updated the marketing site download links and release references to point at v2026.04.1.

Previous Release: v2026.03.11

  • Fixed the macOS DMG packaging regression that caused the bundled backend to exit on first launch.

Distribution Notes

Unsigned DMG (Apple Gatekeeper)

As of April 1, 2026, the MimikaStudio DMG is not yet signed/notarized by Apple.
macOS may block first launch until you explicitly allow it in security settings.

  1. Open the DMG and drag MimikaStudio.app to Applications.
  2. In Applications, right-click MimikaStudio.app and select Open.
  3. Click Open in the warning dialog.
  4. If macOS still blocks launch, go to: System Settings -> Privacy & Security -> Open Anyway (for MimikaStudio), then confirm with password/Touch ID.
  5. On first launch, wait for the bundled backend to start.
  6. On first use, click Download for required models in-app.
Assets 8
Loading

MimikaStudio v2026.03.11

24 Mar 16:49
@BoltzmannEntropy BoltzmannEntropy
v2026.03.11
2247224

Choose a tag to compare

View all tags
MimikaStudio v2026.03.11

MimikaStudio v2026.03.11 Release Notes

Release Date: March 24, 2026
Platform: macOS (Apple Silicon)

What's New In v2026.03.11

  • Fixed the packaged macOS app bundle so the bundled backend starts correctly on first launch from the release DMG.
  • Rebuilt the embedded backend/python packaging flow so the final .app is re-signed and verified after resources are injected.
  • Kept the existing unsigned-DMG guidance in place for Gatekeeper, while removing the broken sealed-bundle state that caused backend exit-on-start.

User Impact

  • New DMG installs should no longer fail with Backend process exited before becoming healthy on first launch.
  • The bundled backend now survives the initial startup path instead of exiting with code 1 because of an invalid packaged app seal.
  • Users still need the documented first-open Gatekeeper/quarantine workaround until the app is signed and notarized.

Technical Notes

  • Updated scripts/build_dmg.sh to strip copied extended attributes from the finished app bundle.
  • Added a final app-bundle codesign --deep pass after embedding backend resources.
  • Added strict post-build signature verification so broken release bundles fail during packaging instead of after shipping.

Previous Release: v2026.03.10

  • Added desktop drag-and-drop import to the Audiobooks workflow, plus resizable/collapsible panels for Documents, Generated Audiobooks, and Upload Voice management.

Distribution Notes

Unsigned DMG (Apple Gatekeeper)

As of March 24, 2026, the MimikaStudio DMG is not yet signed/notarized by Apple.
macOS may block first launch until you explicitly allow it in security settings.

  1. Open the DMG and drag MimikaStudio.app to Applications.
  2. In Applications, right-click MimikaStudio.app and select Open.
  3. Click Open in the warning dialog.
  4. If macOS still blocks launch, go to: System Settings -> Privacy & Security -> Open Anyway (for MimikaStudio), then confirm with password/Touch ID.
  5. On first launch, wait for the bundled backend to start.
  6. On first use, click Download for required models in-app.
Loading

MimikaStudio v2026.03.10

22 Mar 19:16
@BoltzmannEntropy BoltzmannEntropy
v2026.03.10
b9b4b81

Choose a tag to compare

View all tags
MimikaStudio v2026.03.10

MimikaStudio v2026.03.10 Release Notes

Release Date: March 22, 2026
Platform: macOS (Apple Silicon)

What's New In v2026.03.10

  • Added desktop drag-and-drop import to the Audiobooks document library for PDF, EPUB, DOCX, HTML, RTF, ODT, DOC, TXT, and Markdown files.
  • Added explicit in-app drop messaging so the Audiobooks screen now clearly signals where to drop supported documents.
  • Made the Audiobooks Documents pane collapsible and draggable to resize.
  • Made the Generated Audiobooks pane collapsible and draggable to resize.
  • Split Voice Prompts upload into a dedicated Upload Voice pane that is now collapsible and draggable to resize.

User Impact

  • You can now drag supported book/document files straight into the Audiobooks workflow instead of importing them only through the picker.
  • Long-form generation controls are easier to manage because the Documents and Generated Audiobooks areas can be collapsed when you want more room for preview and settings.
  • Voice Prompt Management has a clearer layout: upload stays in its own pane, while search/filter/audition work remains focused in the table below.

Technical Notes

  • Added the desktop_drop Flutter dependency and updated the macOS plugin registrant for native file-drop handling.
  • Refactored flutter_app/lib/screens/audiobook_screen.dart to support dropped local files, explicit drop states, and split-pane resizing/collapse controls.
  • Refactored flutter_app/lib/screens/voice_prompt_management_screen.dart to separate upload controls into their own resizable pane above the shared voice library table.

Previous Release: v2026.03.9

  • Fixed long-form audiobook generation failures for oversized extracted texts and replaced the transient audiobook start-failure snackbar with a readable dialog.

Distribution Notes

Unsigned DMG (Apple Gatekeeper)

As of March 22, 2026, the MimikaStudio DMG is not yet signed/notarized by Apple.
macOS may block first launch until you explicitly allow it in security settings.

  1. Open the DMG and drag MimikaStudio.app to Applications.
  2. In Applications, right-click MimikaStudio.app and select Open.
  3. Click Open in the warning dialog.
  4. If macOS still blocks launch, go to: System Settings -> Privacy & Security -> Open Anyway (for MimikaStudio), then confirm with password/Touch ID.
  5. On first launch, wait for the bundled backend to start.
  6. On first use, click Download for required models in-app.
Loading

MimikaStudio v2026.03.9

21 Mar 01:49
@BoltzmannEntropy BoltzmannEntropy
v2026.03.9
7af1ee9

Choose a tag to compare

View all tags
MimikaStudio v2026.03.9

MimikaStudio v2026.03.9 Release Notes

Release Date: March 21, 2026
Platform: macOS (Apple Silicon)

What's Fixed In v2026.03.9

  • Fixed long-form audiobook generation failures for very large EPUB and document inputs that exceeded spaCy's default 1,000,000-character limit during sentence splitting.
  • Added a safe fallback so oversized audiobook texts use regex sentence splitting instead of aborting before chunking begins.
  • Replaced the transient audiobook start-failure snackbar with a readable dialog so users can inspect the full backend error message.

User Impact

  • Large books that previously failed with errors like Text Length 1015801 exceeds 1000000 can now proceed into audiobook chunking and queueing.
  • Users now get a persistent error dialog when audiobook startup fails, instead of a brief bottom toast that disappears before the message can be read.
  • EPUB chapter extraction remains in place; the failure was in full-document sentence splitting after extraction, not in the chapter parser itself.

Technical Notes

  • Updated backend/tts/text_chunking.py to detect texts beyond spaCy's max_length and fall back to the regex sentence splitter.
  • Added a second fallback path for spaCy E088 errors so oversized texts continue through audiobook chunking without raising.
  • Added backend regression tests covering both oversized-input fallback and explicit E088 fallback behavior.
  • Updated the Flutter audiobook screen to present startup errors in a modal dialog with selectable text.

Previous Release: v2026.03.8

  • Restored Hebrew Dicta support for the MLX Chatterbox runtime.

Distribution Notes

Unsigned DMG (Apple Gatekeeper)

As of March 21, 2026, the MimikaStudio DMG is not yet signed/notarized by Apple.
macOS may block first launch until you explicitly allow it in security settings.

  1. Open the DMG and drag MimikaStudio.app to Applications.
  2. In Applications, right-click MimikaStudio.app and select Open.
  3. Click Open in the warning dialog.
  4. If macOS still blocks launch, go to: System Settings -> Privacy & Security -> Open Anyway (for MimikaStudio), then confirm with password/Touch ID.
  5. On first launch, wait for the bundled backend to start.
  6. On first use, click Download for required models in-app.
Loading

MimikaStudio 2026.03.8

18 Mar 14:59
@BoltzmannEntropy BoltzmannEntropy
v2026.03.8
af3ec1f

Choose a tag to compare

View all tags
MimikaStudio 2026.03.8

MimikaStudio v2026.03.8 Release Notes

Release Date: March 18, 2026
Platform: macOS (Apple Silicon)

What's Fixed In v2026.03.8

  • Restored Hebrew Dicta diacritization for Chatterbox Multilingual after the MLX backend migration.
  • Fixed the regression where Hebrew Chatterbox generation could silently fall back to a broken tokenizer path and log:
    Dicta.__init__() missing 1 required positional argument: 'model_path'
  • Rewired MimikaStudio's Chatterbox engine to preload the Dicta ONNX model into the tokenizer module actually used by the MLX runtime before model load.

User Impact

  • Hebrew Chatterbox smoke tests now run without the Dicta initialization warning.
  • Full Hebrew voice-clone renders complete successfully again on the patched MLX path.
  • Existing Hebrew Dicta installs are now picked up from:
    • DICTA_ONNX_MODEL_PATH
    • bundled app model path
    • Mimika runtime data model path

Technical Notes

  • Updated the backend Chatterbox engine to resolve an explicit Dicta ONNX path instead of relying on a zero-argument Dicta() constructor.
  • Patched both supported tokenizer import paths so Hebrew preprocessing is applied consistently across the MLX-backed Chatterbox runtime.
  • Verified generation with:
    • a short Hebrew smoke test
    • a full Hebrew parashat_hashavua.txt render using a custom uploaded reference voice

Previous Release: v2026.03.7

  • Sandboxed release entitlement update and release-build hardening.

Distribution Notes

Unsigned DMG (Apple Gatekeeper)

As of March 18, 2026, the MimikaStudio DMG is not yet signed/notarized by Apple.
macOS may block first launch until you explicitly allow it in security settings.

  1. Open the DMG and drag MimikaStudio.app to Applications.
  2. In Applications, right-click MimikaStudio.app and select Open.
  3. Click Open in the warning dialog.
  4. If macOS still blocks launch, go to: System Settings -> Privacy & Security -> Open Anyway (for MimikaStudio), then confirm with password/Touch ID.
  5. On first launch, wait for the bundled backend to start.
  6. On first use, click Download for required models in-app.
Loading