MaRi23333/dsh-fish-tts 预览 preview

MaRi23333/dsh-fish-tts

Plugin插件 Native原生 ⭐ 2 MIT Vision & Media视觉与多媒体

Fish Audio TTS插件,适用于DeepSeek Harness——支持逐条朗读、自动朗读开关、自带Key、第三方非官方。语音合成插件:逐条朗读、自动朗读;仅支持Fish Audio API,需自备Key;第三方非官方。

Project Overview项目介绍

dsh-fish-tts is a native third-party text-to-speech plugin built exclusively for the DeepSeek Harness (DSH) web GUI. It only integrates with the Fish Audio TTS API, and requires users to bring their own Fish Audio API key to use the service. The plugin adds a one-click read-aloud button to every assistant response, a toggle for auto-reading new replies, and a dedicated settings page for configuring models, voices, API keys, and proxies. To install the plugin, you can run a single npx command npx @deepseek-ai/dsh plugin --profile web add dsh-fish-tts via the DSH CLI, then restart the DSH web service and refresh your browser page to activate it.

This plugin is designed for DSH users who already have a Fish Audio account and want to use custom voices, including voices cloned by the user themselves. After installation, the first step to use the plugin is to open the DSH settings menu, navigate to the newly added Voice (Fish TTS) section, and enter your Fish Audio API key and your desired voice reference ID. You can then click the test listen button to verify that your configuration, network connection, and API access all work correctly. Once configured, you can leave auto-read enabled to have every new assistant response read aloud automatically, or manually click the read button on individual messages.

dsh-fish-tts is released under the open-source MIT license, and requires Node.js version 22 or higher to build and run locally. All API keys are stored locally encrypted with AES-256-GCM, and the plugin blocks all non-local access to its endpoints to prevent unauthorized access. The plugin has a 12000 character limit per single synthesis request, does not support proxies with username and password authentication, and caches up to 200 recent synthesis results in memory, which are cleared on DSH restart. Users are required to comply with Fish Audio’s terms of service when using this plugin, and the plugin is an independent community project not affiliated with either DeepSeek or Fish Audio.

dsh-fish-tts 是专为 DeepSeek Harness(DSH)Web 界面开发的第三方原生 TTS 插件,仅对接 Fish Audio 语音合成 API,可为 DSH 中助手的每条回复添加一键朗读功能,同时支持自动朗读新回复、配置自定义模型与音色。用户可通过 DSH 命令行一键安装,安装后重启 DSH 服务刷新页面即可启用插件。

本插件适合已经拥有 Fish Audio 账号,想要使用自定义音色(包括自行克隆的音色)的 DSH 用户使用。使用前需先在插件设置页填入 Fish Audio API Key、目标音色的 reference_id,配置完成后可点击试听验证配置是否正确。开启自动朗读后,每次助手生成新回复都会自动朗读,无需手动点击。

本插件采用 MIT 许可开源,要求 Node.js 版本不低于 22,API Key 会通过 AES-256-GCM 加密存储在本地,不会泄露到外部。单条合成请求文本上限为 12000 字符,不支持带认证信息的代理,仅允许本地访问接口,杜绝远程访问风险。用户需要自行遵守 Fish Audio 的服务条款。

Pre-install check安装前体检Compatibility · Security兼容性 · 安全性 1 warning1 项注意
  • Only 2 stars - very few users, little community feedback星标只有 2,几乎没人在用,遇到问题缺少社区反馈
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

npx @deepseek-ai/dsh plugin --profile web add dsh-fish-tts

把 MaRi23333/dsh-fish-tts 加入你的 DSH 配置(web profile)即可启用。

READMEREADME

dsh-fish-tts

中文 | English

dsh-fish-tts — DeepSeek Harness 语音合成插件(仅支持 Fish Audio API)

CI License: MIT DeepSeek Harness 0.2.0-rc.2

English: dsh-fish-tts is a third-party TTS plugin for the DeepSeek Harness Web GUI — Fish Audio API only, bring your own key: a per-message "Read aloud" action, an auto-read toggle in the composer, and a settings page for model / voice / encrypted API key / proxy. Bilingual UI (zh/en). See README.en.md for the full English version.

简介

面向 DeepSeek Harness(DSH)Web GUI 的第三方语音合成(TTS)插件:为每条助手回复提供一键朗读,输入栏带自动朗读开关,模型 / 音色 / 加密 Key / 代理均可配置。仅支持 Fish Audio API,需自备 API Key。 使用你自己的 Fish 音色(reference_id,含自行克隆的音色)——朗读、试听都走你配置的语音。

30 秒对比:与常见 Edge TTS 朗读插件

对比项 dsh-fish-tts(本插件) 常见 Edge TTS 朗读插件
合成引擎 Fish Audio 官方 API(仅支持;需自备 Key) 微软 Edge 内置语音
音色 你自己的音色 reference_id(含自行克隆的音色,须有使用权限) Edge 内置固定音色库
API Key 必须自备(设置页 AES-256-GCM 加密保存) 无需
适合谁 已有 Fish Audio 账号、想要自己或克隆音色的用户 想免费快速试用、固定音色即可的用户

功能

  • 朗读按钮:每条定稿的助手消息操作条里有一个「朗读 / Read aloud」按钮(图标与原生操作条图标同风格),点击合成并播放该条回复;播放中再次点击该消息的按钮即中断停止(不会从头重播),点击其他消息的按钮则直接切换播放,合成等待中再点一次可取消。markdown 会被清洗:路径 / URL / 长编号 / 代码块不会读出来。
  • 自动朗读:输入框工具行的小喇叭开关(与设置页同步),开启后自动朗读页面加载后产生的新回复。
  • 设置页(Settings → 语音朗读 / Voice (Fish TTS)):
    • TTS 模型(下拉建议 + 手动输入,支持 s2.1-pro-free / s2.1-pro / s2-pro 等,保存后立即生效;默认 s2.1-pro-free)
    • 音色 reference_id(必填:音色是个人数据,插件不提供默认音色;未填写时合成会被拒绝并提示)
    • API Key(AES-256-GCM 加密存储在本机 $DSH_HOME/fish-tts/settings.json,密钥文件 key.bin 自动生成并在 Windows 上收紧 ACL;Key 不会出现在任何 GET 响应、日志或仓库中)
    • HTTP 代理(例如 http://127.0.0.1:7890,直连不通时可填)
    • 试听按钮、自动朗读开关、音量滑条(默认 60%)、倍速滑条(0.5–2.0×,不变调;浏览器不支持时固定 1×)
  • 双语界面(中文 / English,跟随 DSH 语言设置;en 词条与 zh 完全对齐)。

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-myskin 下一个 Next dsh-animation-studio →