DeepSeek HarnessHub
Submit plugin

molanxuan-del/DSH-VoxPolish ↗★ 3

dsh-voice-polish

「声琢」VoxPolish — 把说出来的话打磨成清晰表达：原话/整理后双通道，先说后改再发送，全程不碰输入框。Voice input with an expression coach for DeepSeek Harness; local SenseVoice ASR, no extra model download. 适合需要高效语音输入、不愿手动修改草稿的DSH用户，无需额外下载模型。

Other extensions
Package
dsh-voice-polish
Compatibility
Unverified
Version
1.0.0
License
MIT
Last updated
Oct 15, 2026

Other repositories with this package name

Install

$npx -p @deepseek-ai/dsh dsh plugin --profile web add github:molanxuan-del/DSH-VoxPolish

README

Read the full README ↗

配置

默认不写死模型，整理用哪条路由按优先级决定：

优先级来源说明
1面板里选的（⚙ 整理模型）存 ~/.dsh/voice-polish/config.json，可随时切回「跟随会话默认」
2插件配置 provider / model想固定一条便宜路由（不占主力模型额度）时写在这里
3会话默认模型兜底，跟随你在 DSH 里的设置
- insert:
    - id: voice-polish
      name: 'dsh-voice-polish'
      config:
        provider: deepseek-official
        model: deepseek-flash
        temperature: 0.2
        maxTokens: 2400

整理是改写任务不是推理任务，便宜快的模型就够。