WizisCool/dsh-ears 预览 preview

WizisCool/dsh-ears

Plugin插件 Native原生 ⭐ 21 MIT Vision & Media视觉与多媒体

DeepSeek Harness(dsh)的语音输入插件:麦克风 → 转写 → 可选润色 → 可编辑草稿

Project Overview项目介绍

dsh-ears is a native ASR (automatic speech recognition) plugin built exclusively for DeepSeek Harness (DSH), adding voice-to-text input and LLM-based text polishing capabilities directly to the DSH web interface. It supports a wide range of ASR backends, from browser-native Web Speech and locally running Whisper powered by whisper.cpp to over 10 popular cloud ASR services including Groq, Deepgram, Tencent Cloud, and Alibaba Cloud Bailian. It also accepts custom OpenAI-compatible ASR endpoints, letting users connect to any compatible ASR service they prefer to use with DSH.

After installation completes, a microphone icon will appear next to the DSH web input bar. Users can click this icon or press a customizable keyboard shortcut (default Ctrl+Shift+Space) to start and stop recording. When recording stops, the plugin automatically transcribes the audio to text, and if polishing is enabled, it will send the raw transcription to an LLM to clean up filler words and fix errors before placing the final result in the DSH input draft. This plugin is designed for any DSH user who wants to speed up their workflow with voice input.

To install and run dsh-ears, you need DSH version 0.1.2-rc.1 or newer, plus Node.js version 22.19.0 or higher to build and run the native dependencies for local Whisper. The simplest installation method is via the DSH CLI with the command dsh plugin --profile web add dsh-ears, and it can also be built and installed directly from source code. The plugin is released under the permissive open source MIT license, and it has several documented limits, such as a 24MB size cap for local Whisper transcription and Web Speech relying on third-party browser services.

dsh-ears 是专为 DeepSeek Harness (DSH) 开发的原生语音输入插件,核心功能是为 DSH 提供语音转文字(ASR)输入与 LLM 润色整理能力。它支持多种识别后端,包括浏览器自带的 Web Speech、本地运行的 Whisper,以及 Groq、Deepgram、阿里云百炼等十余家主流云 ASR 服务,还兼容自定义 OpenAI 兼容 ASR 端点。

安装完成后,DSH Web 界面输入框右侧会新增麦克风图标,用户可点击图标或使用自定义快捷键开始、停止录音。停止录音后插件会自动转写语音,开启润色功能后还会调用 LLM 整理内容,结果写入输入草稿,不会自动发送,方便用户检查编辑。它面向所有需要通过语音输入提升 DSH 使用效率的用户。

该插件要求 DSH 版本不低于 0.1.2-rc.1,依赖 Node.js 22.19.0 或更高版本,可通过 npm 一键安装,也支持从源码编译安装。它采用 MIT 许可证开源,存在部分后端限制,比如 Web Speech 需依赖浏览器厂商服务,本地 Whisper 单次音频不超过 24MB。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 note1 项提示
  • 21 stars - an early-stage project星标 21,属于早期项目
DSH walks through these 9 checksDSH 会逐条核对这 9 项

Compatibility兼容性

  • DSH, Node, OS and profile requirementsDSH 版本 / Node 版本 / 操作系统 / profile 是否满足要求
  • External dependencies and runtimes (Electron / Python / Docker, ...)外部依赖与运行时(Electron / Python / Docker 等)是否齐备
  • Conflicts with installed plugins: command names, skill / tool names, ports, duplicate MCP registration与已装插件是否冲突:命令名、skill / tool 重名、端口占用、重复 MCP 注册

Security安全性

  • Repo matches the facts registered here; archived or abandoned?仓库是否与页面登记一致,是否归档或长期停更
  • Safety of preinstall / install / postinstall and install.sh / setup.ps1preinstall / install / postinstall 与 install.sh、setup.ps1 是否安全
  • curl|bash, download-then-execute, obfuscation, unrelated domains → stop immediatelycurl|bash、下载即执行、混淆代码、无关域名 → 立刻停止
  • Typosquatting or unmaintained packages among the new dependencies新增依赖里有没有 typosquatting 或无人维护的包
  • Requested permissions vs. what the feature actually needs申请了哪些权限、是否超出功能所需(filesystem / network / shell / clipboard)
  • Any sudo / admin requirement, plus uninstall and rollback是否要求 sudo / 管理员权限,以及卸载与回滚方式

Anything uncertain must be marked unknown with a note on how to confirm it. This site's signal screen is a static snapshot, not a security audit.拿不准的必须标「未知」并说明要我怎么确认。本站的信号筛查是静态快照,不能替代安全审计。

Or use CLI install (for developers)或使用命令行安装(适合开发者)

CLI Install命令行安装

dsh plugin --profile web add dsh-ears

把 WizisCool/dsh-ears 加入你的 DSH 配置(web profile)即可启用。

READMEREADME

dsh-ears

dsh-ears

一款支持润色整理的 DeepSeek Harness 语音输入插件。

简体中文 · English

dsh >= 0.2.0-rc.1 npm version npm downloads MIT

https://github.com/user-attachments/assets/1363768e-a393-44bd-a008-1ce2055cac41


dsh-ears 为 DeepSeek Harness 提供语音输入与 LLM 润色整理能力。支持浏览器 Web Speech、本地 Whisper 以及主流语音识别(ASR)API 服务。

安装

前置依赖:DeepSeek Harness >=0.2.0-rc.1,以及 Node.js ^22.19.0 || >=24.0.0。

通过 npm 安装

Web UI:

dsh plugin --profile web add dsh-ears

桌面端:在插件管理界面中安装。桌面端独占名为 desktop 的 profile,dsh 命令不会管理它,因此没有对应的命令行写法。dsh-ears 不需要单独的桌面端构建,两个界面加载同一个 web 客户端包。

仍在使用 dsh 0.1.x? dsh-ears 0.4.1 要求 dsh >=0.2.0-rc.1,dsh 会直接拒绝在旧版本上安装。请继续使用对应的旧版本线:

# dsh 0.1.2–0.1.6
dsh plugin --profile web add "dsh-ears@<0.4.0"
# dsh 0.1.7(dsh-ears 0.4.0 要求 dsh 0.1.7-rc.2 及以后)
dsh plugin --profile web add "dsh-ears@<0.4.1"
# dsh 0.1.1
dsh plugin --profile web add "dsh-ears@<0.3.0"

Showing the opening section of the README — the full document lives in the repository以上为 README 开头摘要,完整文档在仓库内 · View the full README on GitHub →在 GitHub 查看完整 README →

← 上一个 Prev DSH-Desktop 下一个 Next kiwi →