Skip to content

Releases: shadoprizm/cursay

Release list

Cursay 1.3.0 — Ubuntu preview

Cursay 1.3.0 — Ubuntu preview Pre-release
Pre-release

Choose a tag to compare

View all tags
@shadoprizm shadoprizm released this 12 Oct 02:47
ubuntu-v1.3.0
c6f58f3
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Ubuntu preview of Cursay 1.3.0.

Adds opt-in future-capture Memory sync, private capture controls, approved vocabulary hints, and a link to the browser workspace. Preserves automatic Ctrl+Shift+V for supported terminals, with a manual paste override. The installer now bundles Ubuntu's secure credential helper for account linking.

Validated with 78 Ubuntu tests, Python lint, shell checks, an isolated GTK smoke run, and fresh-install verification. GitHub Linux, Mac build, and Windows build checks passed.

Memory still requires separate account consent and production storage-readiness checks. This preview does not import old history. Signed Mac packages and installed Windows acceptance are not included in this Ubuntu preview.

Download the archive and checksum, verify the checksum, extract the archive, and run ./install.sh. Use --skip-backend when preserving an existing external speech service.

Loading

Cursay 1.2.1

Cursay 1.2.1 Latest
Latest

Choose a tag to compare

View all tags
@shadoprizm shadoprizm released this 22 Sep 18:21
v1.2.1
628dc1a

1.2.1 - 2026-09-22

Added

  • Added a fully pinned, hash-verified lockfile for the bundled transcription backend.

Changed

  • The installer now enforces the locked backend dependency set.
  • Linked the canonical Cursay website from the README and GitHub repository homepage.
Assets 4
Loading

Cursay 1.2.0

Cursay 1.2.0

Choose a tag to compare

View all tags
@shadoprizm shadoprizm released this 22 Sep 17:59
v1.2.0
3f68e7b

1.2.0 - 2026-09-22

Added

  • Added a native macOS developer preview with push-to-talk dictation, automatic paste, local history, and a menu-bar controller.
  • Added Prompt writing mode for turning spoken intent into a ready-to-paste AI prompt.
  • Added an Ubuntu in-app updater that verifies versioned release archives against their published SHA-256 checksums.
  • Added optional Smart Polish rewriting with distinct Professional and Casual styles.
  • Added visible fallback messaging when Smart Polish is unavailable.
  • Added local transcription-cost insights with provider/model breakdowns.
  • Added support for exact provider-reported request costs when compatible endpoints return them.
  • Added duration-based xAI REST speech-to-text estimates using its published rate.

Changed

  • Clarified that Code and Raw modes never use Smart Polish.
  • Improved the Dictate screen's explanation of each writing mode.
  • Databases are migrated automatically to store optional provider-reported costs.
  • Installing with --skip-backend now removes a previously enabled bundled STT service.
Loading