PerryLink/dsh-talk

DeepSeek Harness的语音优先会话循环:一个带浏览器/本地语音转文字(Web Speech、FunASR、whisper.cpp)的作曲家麦克风按钮,一个用于文本转语音回复的讲话工具(浏览器、edge-tts、piper),带静音功能的事件播报,以及说话即打断。

Project Overview项目介绍

dsh-talk is a native voice interaction plugin built exclusively for DeepSeek Harness (DSH). It adds a full bidirectional voice loop to DSH, allowing users to speak to the agent directly via microphone and listen to the agent’s spoken replies. The plugin supports multiple speech-to-text (STT) and text-to-speech (TTS) engines, and can be installed through DSH’s plugin manager via git, npm, or tarball distribution channels. To install from the official npm release channel, run dsh plugin --profile web add dsh-talk after you have installed the global dsh1024 tool. For users who want the latest unreleased development version, installing via the git channel is the recommended option.

After installation, you can press the microphone button that appears next to DSH’s input composer to start recording your voice. The plugin includes a convenient speak-to-interrupt feature, which automatically pauses any ongoing audio playback as soon as you start speaking. You can configure optional toggle hotkeys, set a maximum recording duration, and adjust all settings including selecting your preferred STT and TTS engine and language directly from a dedicated settings tab in DSH. All configuration changes are saved as append-only profile patches with automatic backups, so you never have to worry about losing your custom settings.

The plugin requires Node.js version 22.19.0 or newer, and works best on Chrome or Edge desktop browsers due to their full support for Web Speech API and MediaRecorder. It is released under the open-source Apache 2.0 license, so it is free to use, modify, and distribute per the license terms. After installation, you can run dsh --profile web --dump-config | grep -A2 'id: talk' to verify that the plugin has been loaded correctly. If you encounter an ERR_PNPM_IGNORED_BUILDS error during installation, you can fix it by adding allowBuilds: { esbuild: true } to your pnpm-workspace.yaml file as instructed.

这是专门为DeepSeek Harness(DSH)开发的原生语音交互插件,为DSH提供完整的双向语音交互闭环,用户可以直接对着麦克风说话,也能听到DSH用语音回复内容。它支持多种语音识别(STT)和语音合成(TTS)引擎,可通过DSH自带的插件管理工具,从git、npm、压缩包三种渠道完成安装。

安装完成后,用户可以方便地点击输入框旁的麦克风按钮开始录音说话,插件支持说话打断功能,用户一开始说话就会自动中断正在播放的音频。它支持自定义录音时长、可配置开关麦克风的快捷键,所有配置项都会以配置补丁的形式保存并自带备份,还提供专门的设置面板方便选择引擎和语言。

该插件要求Node版本不低于22.19.0,目前在Chrome和Edge浏览器上体验最佳,兼容DSH 0.1.5-rc.2及以上版本,采用Apache 2.0开源许可证免费使用。安装后可通过dsh命令检查插件是否正确加载,若遇到pnpm编译错误可按照官方提示修改配置文件解决。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 note1 项提示
  • 16 stars - an early-stage project星标 16,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-talk

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

READMEREADME

🎙️ dsh-talk

  • 1024 store channel: npm i -g dsh1024 once, then dsh1024 plugin --profile web add dsh-talk (counts toward the deepseek1024.com install ranking). Gitee dshfind OpenSSF Scorecard

Voice-first session loop for DeepSeek Harness: talk to it, hear it answer.

Press the mic, speak, and the reply is spoken back — with speak-to-interrupt.

License DSH plugin dsh-doctor DSH Market Node CI Version npm version npm downloads

English · 简体中文 · Español · Português · हिन्दी


⭐ 如果它帮到了你

这个插件是 DSH 插件家族的一员(40+ 个,全部 Apache-2.0)。如果你在用,给个 star —— 它不会解锁任何功能,但会让下一个人在搜索里更容易找到它。

English: part of a 40+ plugin family for DeepSeek Harness. If it is useful, a star helps the next person find it — nothing is gated behind it.

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-plugin-message-edit 下一个 Next deepseek-harness-zh-cn →