Skip to content

Releases: yuxino/mimi

mimi v1.3.3

Choose a tag to compare

@github-actions github-actions released this 29 Aug 15:09

Developer-installable packages for macOS (Apple Silicon) and Windows x64. The macOS package is ad-hoc signed and not notarized; on first launch, choose Open Anyway in System Settings → Privacy & Security if macOS blocks it. Updates may require approving Screen & System Audio Recording and Keychain access again. The Windows MSI and NSIS installers are an unsigned preview without Authenticode signing, so Windows Defender SmartScreen may warn; real-device acceptance is still incomplete.

mimi v1.3.2

Choose a tag to compare

@github-actions github-actions released this 28 Aug 15:22

Developer-installable packages for macOS (Apple Silicon) and Windows. The macOS package is ad-hoc signed and not notarized; on first launch, choose Open Anyway in System Settings → Privacy & Security if macOS blocks it. Updates may require approving Screen & System Audio Recording and Keychain access again.

mimi v1.3.1

Choose a tag to compare

@github-actions github-actions released this 27 Aug 13:41

Developer-installable packages for macOS (Apple Silicon) and Windows. The macOS package is ad-hoc signed and not notarized; on first launch, choose Open Anyway in System Settings → Privacy & Security if macOS blocks it. Updates may require approving Screen & System Audio Recording and Keychain access again.

mimi v1.3.0

Choose a tag to compare

@github-actions github-actions released this 24 Aug 14:12

Highlights

  • Added a compact subtitle control island that stays available when the subtitle background is hidden.
  • Redesigned the subtitle control panel with one-click language and mode grids, an explicit background restore control, and scrolling on short screens.
  • Hardened the audio and provider pipelines with bounded queues, generation-owned async work, finite deadlines, and safer recovery.
  • Tightened per-window Tauri permissions, production content security, release permissions, and binary size.

Requirements

  • macOS 13 or later on Apple Silicon.
  • Windows 10 or 11 on x64.

Notes

  • Mimi captures system audio only and does not store audio or subtitle content.
  • These community release builds are not currently Apple-notarized or Windows publisher-signed, so the operating system may show a trust prompt.

mimi v1.2.0

Choose a tag to compare

@github-actions github-actions released this 24 Aug 05:08

See the assets below for installers (macOS .dmg, Windows .msi/.exe).

mimi v1.1.0

Choose a tag to compare

@github-actions github-actions released this 15 Aug 09:04

See the assets below for installers (macOS .dmg, Windows .msi/.exe).

mimi v1.0.1

Choose a tag to compare

@github-actions github-actions released this 15 Aug 08:29

See the assets below for installers (macOS .dmg, Windows .msi/.exe).

mimi v1.0.0

Choose a tag to compare

@github-actions github-actions released this 14 Aug 20:12

See the assets below for installers (macOS .dmg, Windows .msi/.exe).

mimi 0.1.5

Choose a tag to compare

@yuxino yuxino released this 12 Aug 15:47

新增极速翻译模式:

  • 极速:用 qwen-mt-flash 边识别边翻译,译文流式上屏,一句话说完即定稿,最快看到字幕。
  • 翻译模式三档切换:极速 / 低延迟 / 高质量,设置里和字幕窗左上角都能直接切换,正在听时切换会自动重连。
  • 模式选择会被记住,不再每次强制切回高质量。

mimi 0.1.4

Choose a tag to compare

@yuxino yuxino released this 12 Aug 15:32

实时字幕体验与翻译质量更新:

  • 字幕只显示译文,不再先显示原文再跳成译文,避免视觉跳动。
  • 浮窗底部新增居中状态提示:正在聆听 / 正在识别 / 正在翻译 / 已暂停。
  • 翻译准确度优化:正式翻译携带最近 6 句对话上下文(翻译记忆),明确要求模型保持人称、语气、专名一致;自动识别语言时把识别出的语言显式传给翻译模型。