dsh-hold-to-talk
English · 中文
Long-press anywhere on the composer card in DeepSeek Harness to dictate.
Release to transcribe and drop the text into the draft — nothing is sent automatically.
┌──────────────────────────────────────────────┐
│ 按住鼠标 语音输入文字 ← fades in │ ← hover the input box
│ │
│ ⌄ ＋ 权限 计划 [ model ] [ ➤ ] │
└──────────────────────────────────────────────┘
hold ≈0.35 s ↓
┌──────────────────────────────────────────────┐
│ ▁▃▅▇▅▃▁ 正在聆听 · 松开完成 │ ← release to transcribe
│ Esc 取消 · 上滑取消 │
└──────────────────────────────────────────────┘
Requirements — read this first
This plugin does not ship a speech recogniser. It reuses the one DeepSeek Harness already
has, so both of these must hold before it can do anything:
-
The official voice-input bundle is enabled.
It ships with DeepSeek Harness as @deepseek-ai/dsh-experimental-voice-input-bundle
(Plugins → Voice input / 语音输入). It is what mounts the speech remote that this
plugin calls. If you disable it, this plugin silently registers nothing and the composer
keeps its normal behaviour.
-
The recognition models are prepared.
The official bundle transcribes locally with SenseVoice. On first use the Host
downloads the model files (≈ 240 MB) into ~/.dsh/speech-to-text/sensevoice/models
(model.int8.onnx, tokens.txt, silero_vad.onnx).
Open the voice-input bundle page and run Download and prepare once; or simply click
the built-in microphone button once and follow the setup prompt.
The official bundle has no streaming support, so a transcript appears only after you
release. Audio is transient: it never becomes a session event or an attachment, and only
the text you later submit is recorded.
Tested against DeepSeek Harness 0.1.7-rc.2 (the version shipped with the desktop app at
the time of writing). The plugin uses two internal interfaces — the conversation.input.overlay
slot and the speech remote — and is written to degrade instead of throwing when either is
missing.
Install
DSH plugins are installed from a package directory, so clone first:
git clone https://github.com/jryang1997/dsh-hold-to-talk.git
Then pick one route.
A. Through the agent (recommended, works in the Desktop app)
Ask the agent in any session:
Install the bundle at <absolute path to the clone> with plugin_manager
(action: install_bundle).
It runs install_bundle for the current profile and reports application: applied
when the change is live.
B. Through the CLI
dsh plugin --profile <profile> add <absolute path to the clone>
The Desktop application owns its desktop profile exclusively, so use route A there.
C. Verify
Reload the page (Ctrl+R), hover the message box and look for the hint in the lower-right
corner. If it does not appear, open the browser console and check that the host half
activated.
Usage
| Gesture | Result |
|---|
| Hover the input box | The hint fades in over ~0.6 s |
| Press and hold ≈0.35 s without moving | The whole card becomes a recording surface |
| Release | Transcript is inserted at the caret — not sent |
Press Esc while recording or transcribing | Cancel |
| Hold, then swipe up ≥72 px, then release | Cancel (the panel turns red first) |
| A plain click, a drag, or selecting text | Nothing happens — the gesture never arms |
If the draft changed while recognition was running, the transcript is kept in a small
lower-right chip; click it to insert at the current caret.
Tuning
Everything lives at the top of client.js; there is no build step, so edit and reload.
| Constant | Default | Meaning |
|---|
HOLD_MS | 350 | How long the press must stay still |
ARM_TOLERANCE_PX | 10 | Movement that disarms the gesture |
CANCEL_DISTANCE_PX | 72 | Upward travel that arms "release to cancel" |
MIN_SECONDS | 0.35 | Recordings shorter than this are dropped |
NOTICE_MS | 2800 | How long a one-line notice stays |
How it works
| Concern | Mechanism |
|---|
| Where it renders | One entry in the conversation.input.overlay slot — a floating layer inside the composer card |
| Gesture surface | node.closest('[data-composer-card]') from its own node, listening in the capture phase |
| Never disturbing typing | The layer is pointer-events: none; nothing is preventDefaulted before the hold threshold |
| Recording | getUserMedia + MediaRecorder → OfflineAudioContext resample to 16 kHz mono → 44-byte PCM16 WAV |
| Recognition | ctx.remote.speech.transcribe({ audioBase64 }); provider and language come from the Host config |
| Draft insertion | The slot's own inputActions.captureInsertion() / insertText(text, span), guarded by draftRev |
| Styling | Only --dsw-alias-* theme tokens, so light and dark both work |
| Text | Registered through ctx.locale (zh, en) |
Known limitations
- Pointer-only. The long press is the only entry point; there is no keyboard equivalent
yet, so the feature is not reachable by keyboard alone. A focusable trigger is the obvious
next addition.
- Mouse-first. Touch input is untested: on a touch screen a long press also drives text
selection, so the thresholds would probably need tuning there.
- It leans on two internal DSH interfaces — the
conversation.input.overlay slot and the
speech remote — which can change between Harness releases.
Distribution and discovery
DeepSeek Harness has no plugin marketplace and no submission process: the official
repository states that it cannot accept external pull requests at the moment, and the
mechanism it points third-party plugins at is the GitHub topic
dsh-plugin. This repository carries that topic, which
is how it is meant to be found.
install_bundle also accepts an npm package name, so if this plugin is ever published, a user
could install it by name instead of cloning. Publishing to npm would require renaming the
package from @local/... to a scope the publisher owns.
Uninstall
plugin_manager → action: remove_bundle → target: @local/dsh-hold-to-talk
The plugin is installed as a link to the clone, so deleting the clone directory after
removing the bundle is safe.
Privacy and disclaimer
- Audio is captured locally, handed to the Host's speech service over the local API, and
discarded. It never becomes a session event or an attachment.
- If the Host is configured with a cloud speech provider, that audio leaves the machine.
This plugin sends no
providerId, so whichever provider the Host has selected is used; the
default configuration transcribes locally with SenseVoice. Check Settings → Plugins → Voice
input before dictating anything sensitive.
- The plugin makes no network request of its own and stores nothing.
- Unofficial community plugin. Not affiliated with, endorsed by, or supported by DeepSeek;
"DeepSeek Harness" is named only to describe compatibility. The code here is original and
carries no code from the Harness packages.
License
MIT
中文
按住鼠标说话 —— 在 DeepSeek Harness 的输入框里长按鼠标即可语音输入，松开后转写并插入草稿，
不会自动发送。
前置依赖（先读这一节）
本插件不自带语音识别引擎，它复用 DeepSeek Harness 已有的那套。所以下面两条必须成立：
-
官方的语音输入 bundle 处于启用状态。
它是随 DSH 一起发布的 @deepseek-ai/dsh-experimental-voice-input-bundle（设置 → 插件 →
「语音输入」/ Voice input）。本插件调用的 speech 远程命名空间正是由它挂载的。
一旦禁用它，本插件会静默不注册任何东西，输入框行为完全恢复原样。
-
识别模型已经准备好。
官方 bundle 使用 SenseVoice 在本机转写。首次使用需要下载模型文件（约 240 MB）到
~/.dsh/speech-to-text/sensevoice/models（model.int8.onnx、tokens.txt、silero_vad.onnx）。
打开语音输入 bundle 的详情页点一次「下载并准备」，或者直接点一次内置麦克风按钮并按引导操作。
官方 bundle 不支持流式转写，所以文字只会在你松开之后出现。音频是临时的：不会成为 Session
事件或附件，只有你之后正式提交的文字才会被记录。
安装
DSH 插件从包目录安装，先克隆：
git clone https://github.com/jryang1997/dsh-hold-to-talk.git
然后二选一。
方式 A：让 Agent 装（推荐，桌面版用这个）
在任意会话里对 Agent 说：
用 plugin_manager（action: install_bundle）安装 <克隆下来的绝对路径> 这个 bundle。
它会为当前 profile 执行安装，并返回 application: applied 表示已生效。
方式 B：命令行
dsh plugin --profile <profile> add <克隆下来的绝对路径>
桌面版应用独占它的 desktop profile，所以在桌面版里请用方式 A。
方式 C：验证
刷新页面（Ctrl+R），鼠标移到消息输入框上，右下角会出现提示行。
如果没有出现，打开浏览器控制台确认 Host 半是否激活。
使用
| 手势 | 结果 |
|---|
| 鼠标移入输入框 | 提示行约 0.6 秒慢慢浮现 |
| 按住不动约 0.35 秒 | 整张输入框变成录音面板 |
| 松开 | 转写文字插入光标处，不发送 |
录音中或识别中按 Esc | 取消 |
| 按住后上滑 ≥72 px 再松开 | 取消（面板会先变红提示） |
| 单击、拖拽、拖选文字 | 什么都不发生 —— 手势根本不会激活 |
如果识别期间草稿被改动过，转写结果会保留在右下角的小胶囊里，点一下即可插入到当前光标。
调参
全部在 client.js 顶部；没有构建步骤，改完刷新即可。
| 常量 | 默认 | 含义 |
|---|
HOLD_MS | 350 | 按住多久才算语音输入 |
ARM_TOLERANCE_PX | 10 | 超过这个位移就不激活 |
CANCEL_DISTANCE_PX | 72 | 上滑多少像素进入"松手取消" |
MIN_SECONDS | 0.35 | 短于此长度的录音直接丢弃 |
NOTICE_MS | 2800 | 一行提示停留多久 |
实现原理
| 环节 | 机制 |
|---|
| 渲染位置 | conversation.input.overlay 座位的一个条目 —— 输入框卡片内的浮层 |
| 手势面 | 从自身节点 closest('[data-composer-card]') 拿到卡片，捕获阶段监听 |
| 不干扰打字 | 浮层默认 pointer-events: none；达到长按阈值前不做任何 preventDefault |
| 录音 | getUserMedia + MediaRecorder → OfflineAudioContext 重采样到 16 kHz 单声道 → 44 字节 PCM16 WAV |
| 识别 | ctx.remote.speech.transcribe({ audioBase64 })，provider 与语言由 Host 配置决定 |
| 写草稿 | 座位自带的 inputActions.captureInsertion() / insertText(text, span)，带 draftRev 校验 |
| 样式 | 只用 --dsw-alias-* 主题令牌，明暗主题都正常 |
| 文案 | 通过 ctx.locale 注册（zh、en） |
已知限制
- 只能用指针。 长按是唯一入口，目前没有键盘等价操作，纯键盘用户无法触达。加一个可聚焦的
触发按钮是下一步最该做的事。
- 以鼠标为主。 触摸屏未验证：触摸长按同时会驱动文本选择，阈值大概需要另调。
- 依赖 DSH 的两个内部接口 ——
conversation.input.overlay 座位与 speech 远程命名空间 ——
它们可能随 Harness 版本变化。
分发与发现
DeepSeek Harness 没有插件市场，也没有投稿流程：官方仓库明确表示目前不接受外部 PR，
它为第三方插件指出的发现机制是 GitHub topic
dsh-plugin。本仓库打上了这个 topic，这就是它被找到的方式。
install_bundle 也接受 npm 包名，所以如果将来发布到 npm，用户就能直接用包名安装、不必克隆。
发布 npm 需要把包名从 @local/... 改成发布者自己拥有的 scope。
卸载
plugin_manager → action: remove_bundle → target: @local/dsh-hold-to-talk
插件是以链接方式安装的，卸载后删掉克隆目录即可。
隐私与声明
- 音频在本机采集，经本地 API 交给 Host 的语音服务，用完即弃；不会成为 Session 事件或附件。
- 如果 Host 配置的是云端语音服务，音频会离开这台机器。 本插件不指定
providerId，
用的是 Host 已选定的那个 provider —— 默认配置是本机 SenseVoice 本地转写。
在口述敏感内容前，请先确认「设置 → 插件 → 语音输入」里的选择。
- 插件自身不发起任何网络请求，也不存储任何内容。
- 非官方社区插件，与 DeepSeek 无隶属、背书或支持关系；提及「DeepSeek Harness」仅为说明兼容性。
本仓库代码为原创，不含 Harness 各包中的代码。
许可
MIT
