Skip to content

Releases: XueshiQiao/qduo

v26.10.19

Choose a tag to compare

@github-actions github-actions released this 01 Oct 14:24

What's New in 26.10.19

  • Compare, then replace — A new output for actions that rewrite your text: the popup shows your selection above the result, with removed text in red and added text in green, and Replace puts the result in place once you have checked it. Choose it under Result in an action's editor; the built-in Polish uses it on new installs (#12).
  • 3D Glass: a new popup style (new in 26.10.18) — A glass ring with real depth that leans gently toward the pointer, with a group's actions sliding out as a second ring. Clicks and double-clicks still go through its hole to the text underneath. Pick it in Settings → Appearance → Display Style.
  • A lighter capsule bar (new in 26.10.18) — The capsule now sits on the same Liquid Glass as the ring, without the heavy dark edge it had; a very thin outline stays on by default and can be turned off. A hovered button turns its icon and name blue instead of greying out, and icon and text sizes are adjustable.
  • Dividers on the Liquid ring (new in 26.10.18) — Liquid can draw hairline dividers between actions, the same lines its submenu uses. Off by default; turn them on in Settings → Appearance.
  • Restore Defaults for each style (new in 26.10.18) — Each display style has its own Restore Defaults button, which asks first and resets only that style. Liquid and 3D Glass now keep separate settings, so tuning one no longer changes the other.
  • The Wheel style is gone (new in 26.10.18) — Display Style now offers Liquid, 3D Glass and Capsule. If you used Wheel, QDuo switches you to Liquid with dividers on, the closest match.
  • Smaller fixes (new in 26.10.18) — Settings now says so when the first popup shortcut you record is already taken by another app. The config file's schema works again, so editors offer completion and flag mistakes.

26.10.19 更新内容

  • 先对比,再替换 — 给改写类动作新增的一种输出方式:弹窗把原文和结果上下排开,删掉的字标红、新加的字标绿,看过之后再点「替换」写回去。在动作编辑器的「结果」里选择;新安装时,内置的「润色」默认就用它(#12)。
  • 新的弹窗样式:立体玻璃(26.10.18 新增)— 一个有厚度的玻璃圆环,会朝鼠标方向轻轻倾斜;有子菜单的动作会滑出第二圈。圆环中间的空洞照样能点击、双击下面的文字。在「设置 → 外观 → 显示样式」里选择。
  • 胶囊条更轻盈(26.10.18 新增)— 胶囊条换成和圆环一样的 Liquid Glass,去掉了原来那圈又厚又深的边,只留一圈很细的边框(默认开,可以关)。鼠标停在按钮上时,图标和文字变成蓝色,不再整块变灰;图标和文字大小都可以调。
  • 水玻璃可以显示分隔线(26.10.18 新增)— 水玻璃圆环的各个动作之间可以加上细分隔线,和子菜单里的一样。默认不显示,在「设置 → 外观」里打开。
  • 每种样式都能恢复默认(26.10.18 新增)— 每种显示样式都有自己的「恢复默认」按钮,点了会先确认,只恢复这一种样式。水玻璃和立体玻璃的设置也分开了,调一个不会再带动另一个。
  • 去掉了「轮盘」样式(26.10.18 新增)— 显示样式现在是水玻璃、立体玻璃和胶囊三种。原来用轮盘的,会自动换成最接近的水玻璃,并打开分隔线。
  • 小修复(26.10.18 新增)— 第一次录的弹窗快捷键已经被别的 App 占用时,设置里会明确提示。设置文件的格式说明恢复正常,编辑器又能自动补全、提示写错的地方了。

v26.10.18

Choose a tag to compare

@github-actions github-actions released this 01 Oct 09:48

What's New in 26.10.18

  • 3D Glass: a new popup style — A glass ring with real depth that leans gently toward the pointer, with a group's actions sliding out as a second ring. Clicks and double-clicks still go through its hole to the text underneath. Pick it in Settings → Appearance → Display Style.
  • A lighter capsule bar — The capsule now sits on the same Liquid Glass as the ring, without the heavy dark edge it had; a very thin outline stays on by default and can be turned off. A hovered button turns its icon and name blue instead of greying out, and icon and text sizes are adjustable.
  • Dividers on the Liquid ring — Liquid can draw hairline dividers between actions, the same lines its submenu uses. Off by default; turn them on in Settings → Appearance.
  • Restore Defaults for each style — Each display style has its own Restore Defaults button, which asks first and resets only that style. Liquid and 3D Glass now keep separate settings, so tuning one no longer changes the other.
  • The Wheel style is gone — Display Style now offers Liquid, 3D Glass and Capsule. If you used Wheel, QDuo switches you to Liquid with dividers on, the closest match.
  • Smaller fixes — Settings now says so when the first popup shortcut you record is already taken by another app. The config file's schema works again, so editors offer completion and flag mistakes.
  • Open the popup with a shortcut (new in 26.09.16) — Select text, press your shortcut, and the popup opens where the pointer is. Set it in Settings → General → Popup Shortcut. Thanks @yaoxinghuo for #4.
  • A pause stays paused (new in 26.09.16) — Pausing QDuo now survives quitting and reopening it; before, every launch quietly resumed it.

26.10.18 更新内容

  • 新的弹窗样式:立体玻璃 — 一个有厚度的玻璃圆环,会朝鼠标方向轻轻倾斜;有子菜单的动作会滑出第二圈。圆环中间的空洞照样能点击、双击下面的文字。在「设置 → 外观 → 显示样式」里选择。
  • 胶囊条更轻盈 — 胶囊条换成和圆环一样的 Liquid Glass,去掉了原来那圈又厚又深的边,只留一圈很细的边框(默认开,可以关)。鼠标停在按钮上时,图标和文字变成蓝色,不再整块变灰;图标和文字大小都可以调。
  • 水玻璃可以显示分隔线 — 水玻璃圆环的各个动作之间可以加上细分隔线,和子菜单里的一样。默认不显示,在「设置 → 外观」里打开。
  • 每种样式都能恢复默认 — 每种显示样式都有自己的「恢复默认」按钮,点了会先确认,只恢复这一种样式。水玻璃和立体玻璃的设置也分开了,调一个不会再带动另一个。
  • 去掉了「轮盘」样式 — 显示样式现在是水玻璃、立体玻璃和胶囊三种。原来用轮盘的,会自动换成最接近的水玻璃,并打开分隔线。
  • 小修复 — 第一次录的弹窗快捷键已经被别的 App 占用时,设置里会明确提示。设置文件的格式说明恢复正常,编辑器又能自动补全、提示写错的地方了。
  • 用快捷键弹出(26.09.16 新增)— 选中文字后按下快捷键,菜单就在鼠标所在的位置弹出。在「设置 → 通用 → 快捷键弹出」里设置。感谢 @yaoxinghuo 的建议 #4。
  • 暂停后重启仍是暂停(26.09.16 新增)— 暂停 QDuo 之后退出再打开,现在仍保持暂停;之前每次启动都会悄悄恢复。

v26.09.16

Choose a tag to compare

@github-actions github-actions released this 30 Sep 13:29

What's New in 26.09.16

  • Open the popup with a shortcut — Select text, press your shortcut, and the popup opens where the pointer is. Set it in Settings → General → Popup Shortcut. Turning it on pauses the automatic popup, so it opens only when you ask; resume from the sidebar or the menu bar to have both. The shortcut also works in excluded apps and address bars. Thanks @yaoxinghuo for #4.
  • A pause stays paused — Pausing QDuo now survives quitting and reopening it; before, every launch quietly resumed it.
  • Select inside terminal programs (new in 26.09.15) — Selecting text in herdr, tmux with mouse mode, vim and other programs that make their own selection inside a terminal now opens the popup. Works in Ghostty, Terminal, iTerm2, OTTY, WezTerm, kitty, Alacritty and Warp; add or remove terminals in Settings → General → Terminals. Thanks @ylorn for #8.
  • The reading window keeps the text's own layout (new in 26.09.15) — It shows the text with its own line breaks, instead of starting every sentence on a new line.

26.09.16 更新内容

  • 用快捷键弹出 — 选中文字后按下快捷键,菜单就在鼠标所在的位置弹出。在「设置 → 通用 → 快捷键弹出」里设置。打开后会自动暂停选中即弹,只有按快捷键才弹;想两种都要,在侧边栏或菜单栏点「恢复」。在排除的 App 和浏览器地址栏里,快捷键同样有效。感谢 @yaoxinghuo 的建议 #4。
  • 暂停后重启仍是暂停 — 暂停 QDuo 之后退出再打开,现在仍保持暂停;之前每次启动都会悄悄恢复。
  • 终端里的程序选中文字也能弹出(26.09.15 新增)— 在 herdr、开了鼠标模式的 tmux、vim 这类在终端里自己处理选择的程序中选中文字,现在也会弹出快捷环。支持 Ghostty、Terminal、iTerm2、OTTY、WezTerm、kitty、Alacritty 和 Warp,可以在「设置 → 通用 → 终端」里增删。感谢 @ylorn 的反馈 #8。
  • 朗读窗口保留原文的排版(26.09.15 新增)— 按原文自己的换行显示,不再每句话都另起一行。

v26.09.15

Choose a tag to compare

@github-actions github-actions released this 30 Sep 10:20

What's New in 26.09.15

  • Select inside terminal programs — Selecting text in herdr, tmux with mouse mode, vim and other programs that make their own selection inside a terminal now opens the popup. Works in Ghostty, Terminal, iTerm2, OTTY, WezTerm, kitty, Alacritty and Warp; add or remove terminals in Settings → General → Terminals. Thanks @ylorn for #8.
  • The reading window keeps the text's own layout — It shows the text with its own line breaks, instead of starting every sentence on a new line.
  • Translate with macOS's own translator (new in 26.09.14) — A new System Translate action (Settings → Actions → Add from Template → Text) translates on your Mac with Apple's built-in translator: no AI model, no API key, works offline. Pick the language to translate into; the language of the text is detected. Needs macOS 15 or later; the first time a language is used, macOS may ask to download it. Thanks @elviscai on X for #9.
  • ElevenLabs voices (new in 26.09.14) — Speak can now read with ElevenLabs, with words highlighted as they are read. Two models: flash v2.5 (the default) starts in about half a second; v3 sounds richer but takes about two seconds to start. Choose from 21 ready-made voices, each of which also reads Chinese.

26.09.15 更新内容

  • 终端里的程序选中文字也能弹出 — 在 herdr、开了鼠标模式的 tmux、vim 这类在终端里自己处理选择的程序中选中文字,现在也会弹出快捷环。支持 Ghostty、Terminal、iTerm2、OTTY、WezTerm、kitty、Alacritty 和 Warp,可以在「设置 → 通用 → 终端」里增删。感谢 @ylorn 的反馈 #8。
  • 朗读窗口保留原文的排版 — 按原文自己的换行显示,不再每句话都另起一行。
  • 用 macOS 自带的翻译(26.09.14 新增)— 新增「系统翻译」动作(设置 → 动作 → 从模板添加 → 文本):直接在本机用苹果自带的翻译,不用 AI 模型、不用 API Key,断网也能用。自己选要译成的语言,原文语言自动识别。需要 macOS 15 或更高版本;某个语言第一次用时,macOS 可能会要求先下载。感谢 X 上的 @elviscai 提出 #9。
  • 新增 ElevenLabs 声音(26.09.14 新增)— 朗读现在可以用 ElevenLabs,念到哪个词就高亮哪个词。两个模型:flash v2.5(默认)大约半秒出声;v3 音质更好,但要 2 秒左右才出声。有 21 个预置音色可选,都能念中文。

v26.09.14

Choose a tag to compare

@github-actions github-actions released this 30 Sep 08:28

What's New in 26.09.14

  • Translate with macOS's own translator — A new System Translate action (Settings → Actions → Add from Template → Text) translates on your Mac with Apple's built-in translator: no AI model, no API key, works offline. Pick the language to translate into; the language of the text is detected. Needs macOS 15 or later; the first time a language is used, macOS may ask to download it. Thanks @elviscai on X for #9.
  • ElevenLabs voices — Speak can now read with ElevenLabs, with words highlighted as they are read. Two models: flash v2.5 (the default) starts in about half a second; v3 sounds richer but takes about two seconds to start. Choose from 21 ready-made voices, each of which also reads Chinese.

26.09.14 更新内容

  • 用 macOS 自带的翻译 — 新增「系统翻译」动作(设置 → 动作 → 从模板添加 → 文本):直接在本机用苹果自带的翻译,不用 AI 模型、不用 API Key,断网也能用。自己选要译成的语言,原文语言自动识别。需要 macOS 15 或更高版本;某个语言第一次用时,macOS 可能会要求先下载。感谢 X 上的 @elviscai 提出 #9。
  • 新增 ElevenLabs 声音 — 朗读现在可以用 ElevenLabs,念到哪个词就高亮哪个词。两个模型:flash v2.5(默认)大约半秒出声;v3 音质更好,但要 2 秒左右才出声。有 21 个预置音色可选,都能念中文。

v26.09.13

Choose a tag to compare

@github-actions github-actions released this 29 Sep 06:58

What's New in 26.09.13

  • A proper first launch — New installs open the Getting Started guide again (it had been skipped straight to Settings), and start with the Liquid Glass ring. Existing setups keep the style they use.
  • MiniMax voices — Speak can now read with MiniMax's speech-2.8 voices as well as Qwen-Audio: sound starts in about half a second, and the words light up whole as they are read. Pick from 16 ready-made English and Chinese voices, or enter any voice ID.
  • Speech has its own settings page — Readers now live in Settings → Speech: one row each with a play button, click a row to edit its voice, model and speed, and add new ones from a single + menu. Each provider's API key is one row that shows whether it is saved, and a reader whose key is missing says so.
  • Choose how the spoken word is highlighted — In Settings → Appearance → Reading highlight: a rounded pill (the default, now clearer in both light and dark mode), a highlighter stroke, a solid pill with white text, or karaoke, where text not yet read is faded. Each one glides from word to word, and the text itself never moves.
  • Lighter result windows — Results and the reading window use the same see-through Liquid Glass as the ring, with softer text and more even line spacing. A one- or two-line result gets a window that fits it instead of leaving empty rows below.
  • A calmer reading window — Status such as "Preparing" or "Local cache" sits next to the reader's name, so the text no longer jumps when it appears.
  • Open Settings from the popup — Settings → Actions → Add from Template → Automation adds a Settings action; tap it on the ring or capsule to open QDuo's settings.
  • Smaller fixes — Pressing Return in an API key box saves the key; long text in the reader and key windows scrolls instead of being cut off; deleting a reader stops its preview; the settings logo is sharp at every size.

26.09.13 更新内容

  • 新用户的首次启动 — 新安装会重新打开「开始使用」引导(之前会直接跳到设置页),默认使用 Liquid Glass 圆环。已经在用的用户保持原来的样式不变。
  • 新增 MiniMax 声音 — 朗读除了 Qwen-Audio,现在还能用 MiniMax speech-2.8 的声音:大约半秒就出声,高亮按整个单词走。预置 16 个中英文音色,也可以填任意音色 ID。
  • 朗读有了单独的设置页 — 朗读员挪到了「设置 → 朗读」:每个朗读员一行,带试听按钮;点一行就能改音色、模型和语速;新增统一用一个「+」菜单。每家服务商的 API Key 占一行,一眼就能看到有没有保存;缺少密钥的朗读员会直接标出来。
  • 朗读高亮可以选样式 — 在「设置 → 外观 → 朗读高亮」里有四种:圆角胶囊(默认,浅色和深色模式下都更清楚了)、荧光笔、实心胶囊(白字)、卡拉 OK(还没念到的字变淡)。每种都会从一个词平滑地滑到下一个词,文字本身不会跳动。
  • 更轻盈的结果窗口 — 结果窗口和朗读窗口改用和圆环一样通透的液态玻璃,字色更柔和,行距更均匀。只有一两行的结果,窗口会贴合内容,下面不再空一大块。
  • 朗读窗口更安静 — 「正在准备」「本地缓存」这类状态显示在朗读员名字旁边,出现时正文不再上下跳。
  • 从弹框直接打开设置 — 在「设置 → 操作 → 从模板添加 → 自动化」里可以添加「设置」动作,在圆环或胶囊上点一下就打开 QDuo 的设置。
  • 其他小修复 — 在 API Key 输入框里按回车就会保存;朗读员和密钥弹窗里内容多时可以滚动,不会被截掉;删除朗读员时会停掉它的试听;设置页的 Logo 在各种尺寸下都清晰。

v26.09.12

Choose a tag to compare

@github-actions github-actions released this 29 Sep 06:34

What's New in 26.09.12

  • MiniMax voices — Speak can now read with MiniMax's speech-2.8 voices as well as Qwen-Audio: sound starts in about half a second, and the words light up whole as they are read. Pick from 16 ready-made English and Chinese voices, or enter any voice ID.
  • Speech has its own settings page — Readers now live in Settings → Speech: one row each with a play button, click a row to edit its voice, model and speed, and add new ones from a single + menu. Each provider's API key is one row that shows whether it is saved, and a reader whose key is missing says so.
  • Choose how the spoken word is highlighted — In Settings → Appearance → Reading highlight: a rounded pill (the default, now clearer in both light and dark mode), a highlighter stroke, a solid pill with white text, or karaoke, where text not yet read is faded. Each one glides from word to word, and the text itself never moves.
  • Lighter result windows — Results and the reading window use the same see-through Liquid Glass as the ring, with softer text and more even line spacing. A one- or two-line result gets a window that fits it instead of leaving empty rows below.
  • A calmer reading window — Status such as "Preparing" or "Local cache" sits next to the reader's name, so the text no longer jumps when it appears.
  • Open Settings from the popup — Settings → Actions → Add from Template → Automation adds a Settings action; tap it on the ring or capsule to open QDuo's settings.
  • Smaller fixes — Pressing Return in an API key box saves the key; long text in the reader and key windows scrolls instead of being cut off; deleting a reader stops its preview; the settings logo is sharp at every size.

26.09.12 更新内容

  • 新增 MiniMax 声音 — 朗读除了 Qwen-Audio,现在还能用 MiniMax speech-2.8 的声音:大约半秒就出声,高亮按整个单词走。预置 16 个中英文音色,也可以填任意音色 ID。
  • 朗读有了单独的设置页 — 朗读员挪到了「设置 → 朗读」:每个朗读员一行,带试听按钮;点一行就能改音色、模型和语速;新增统一用一个「+」菜单。每家服务商的 API Key 占一行,一眼就能看到有没有保存;缺少密钥的朗读员会直接标出来。
  • 朗读高亮可以选样式 — 在「设置 → 外观 → 朗读高亮」里有四种:圆角胶囊(默认,浅色和深色模式下都更清楚了)、荧光笔、实心胶囊(白字)、卡拉 OK(还没念到的字变淡)。每种都会从一个词平滑地滑到下一个词,文字本身不会跳动。
  • 更轻盈的结果窗口 — 结果窗口和朗读窗口改用和圆环一样通透的液态玻璃,字色更柔和,行距更均匀。只有一两行的结果,窗口会贴合内容,下面不再空一大块。
  • 朗读窗口更安静 — 「正在准备」「本地缓存」这类状态显示在朗读员名字旁边,出现时正文不再上下跳。
  • 从弹框直接打开设置 — 在「设置 → 操作 → 从模板添加 → 自动化」里可以添加「设置」动作,在圆环或胶囊上点一下就打开 QDuo 的设置。
  • 其他小修复 — 在 API Key 输入框里按回车就会保存;朗读员和密钥弹窗里内容多时可以滚动,不会被截掉;删除朗读员时会停掉它的试听;设置页的 Logo 在各种尺寸下都清晰。

v26.09.10

Choose a tag to compare

@github-actions github-actions released this 29 Sep 01:13

What's New in 26.09.10

  • Read aloud with a natural cloud voice — Speak can now use Alibaba's Qwen-Audio voices besides the system voice. Add readers in Settings → AI Models → Speech (voice, speed, your own API key), pick a default, or choose one per Speak action.
  • Follow along as it reads — Speak opens a window with the whole text and highlights each word as it is spoken, with pause, read-again and copy. A text read before replays instantly from a local cache instead of the network.
  • No popup in the address bar — Selecting in Chrome's or Safari's address bar (and other Chrome-based browsers) no longer opens QDuo; turn it back on in Settings → General. Thanks @Laffery for #6.
  • Paste works in Settings — ⌘V, ⌘C, ⌘X and ⌘A now work in every settings field, including the API key boxes. Thanks @Laffery for #5.
  • Windows stay while you use them — Copying a result no longer closes its window, and unpinning a window only removes the pin; it closes when you click elsewhere or on its close button.
  • New templates — Settings → Actions → Add from Template → Automation adds "Pause QDuo" to the popup, and "Inspect Element", which shows where the selected text sits in the app — handy when reporting a place QDuo should stay out of.

26.09.10 更新内容

  • 用更自然的云端声音朗读 — 朗读除了系统语音,现在还能用阿里的 Qwen-Audio 声音。在「设置 → AI 模型 → 朗读」里添加朗读员(音色、语速、你自己的 API Key),设一个默认的,也可以给每个朗读动作单独选。
  • 边听边看 — 朗读时弹出窗口显示全文,念到哪个字就高亮哪个字,可以暂停、从头再读、复制。读过的文字再读时直接从本地缓存播放,不用再联网。
  • 地址栏里不再弹出 — 在 Chrome、Safari(以及其他基于 Chrome 的浏览器)的地址栏里选中文字,不会再弹出 QDuo;想要可以在「设置 → 通用」里打开。感谢 @Laffery 的反馈 #6。
  • 设置里可以粘贴了 — 所有设置输入框(包括 API Key)现在都支持 ⌘V、⌘C、⌘X、⌘A。感谢 @Laffery 的反馈 #5。
  • 窗口不会突然关掉 — 复制结果后窗口不再自动关闭;取消钉住只是取消钉住,点别处或点关闭按钮才会关。
  • 新模板 — 在「设置 → 操作 → 从模板添加 → 自动化」里可以把「暂停 QDuo」加到弹框上;还有「检查元素」,能看到选中的文字位于 app 界面的什么位置,反馈「这里不该弹出」时很有用。

v26.09.9

Choose a tag to compare

@github-actions github-actions released this 28 Sep 09:16

What's New in 26.09.9

  • Pause QDuo — Pause the popup from the menu bar or with the pause button at the bottom of the Settings sidebar, and resume it the same way. While paused, selecting text opens nothing and the menu bar icon shows a pause mark; the screenshot OCR hotkey keeps working, and the pause lasts across restarts.

26.09.9 更新内容

  • 可以暂停 QDuo — 在菜单栏菜单里,或者点设置窗口侧边栏左下角的暂停按钮,就能暂停弹框,恢复也在同一个地方。暂停期间选中文字不会弹出,菜单栏图标里会出现暂停符号;截图 OCR 快捷键照常能用,退出重开后仍然保持暂停。

v26.09.8

Choose a tag to compare

@github-actions github-actions released this 27 Sep 14:38

What's New in 26.09.8

  • Excluded apps stay quiet — Selecting text in an excluded app now also closes a popup left over from an earlier selection, so it can no longer end up on top of that app. App names in the list match regardless of upper or lower case when you edit the config file by hand.

26.09.8 更新内容

  • 屏蔽的 App 更干净 — 在屏蔽的 app 里选中文字时,之前留下的弹框也会一起关掉,不会再盖在这个 app 上面。手动编辑设置文件时,屏蔽列表里的 app 标识不再区分大小写。