forrestahha/dsh-voice-input
30Last commit Aug 14, 2026
dsh-voice-input DSH plugin
Adds a microphone button to the composer input area. It uses the Web Speech API to stream recognition results into the current draft without automatic submission. The plugin requires no external API keys or server-side services.
How to install the dsh-voice-input DSH plugin
dsh plugin --profile web add github:forrestahha/dsh-voice-inputCopying does not run this command. Review the repository and version before installing the dsh-voice-input DSH plugin.
dsh-voice-input DSH plugin data source
dsh-voice-input DSH plugin snapshot date: Aug 16, 2026
discovered
What the dsh-voice-input DSH plugin can do
- Adds a microphone button to the `conversation.input.right` extension slot.
- Supports both standard `SpeechRecognition` and WebKit-prefixed implementations.
- Uses the browser's language setting, falling back to `zh-CN`.
- Preserves natural spacing for Chinese, Japanese, Korean, and Latin text.
- Stops recording when the microphone button is clicked again and aborts on component unload.
- Disables the button in browsers that do not support SpeechRecognition.
Where the dsh-voice-input DSH plugin fits
- Hands-free drafting of long messages or code explanations.
- Dictating technical documentation or meeting notes.
- Accessibility for users with repetitive strain injuries or limited typing ability.
- Quickly capturing ideas during brainstorming sessions.
- Voice input in environments where typing is inconvenient (e.g., standing desk).
Who the dsh-voice-input DSH plugin is for
- DeepSeek Harness Web UI users who prefer voice input over typing.
- Users with temporary or permanent typing difficulties.
dsh-voice-input DSH plugin limitations
- Requires a browser that supports `SpeechRecognition` or `webkitSpeechRecognition` (Chromium-based browsers offer the broadest support).
- Microphone access needs a secure context (HTTPS), except when the server is running on `localhost`.
- The plugin does not store audio, but the browser's Web Speech API implementation may send audio to the browser vendor's speech service – review your browser's privacy policy for sensitive material.
- Does not automatically submit the message; the user must click send manually.
dsh-voice-input DSH plugin: from the repository README
Quoted from the forrestahha/dsh-voice-input README, the upstream source of the dsh-voice-input DSH plugin. Copyright remains with the original authors.
English | [简体中文](README.zh-CN.md) Voice-to-text input for the [DeepSeek Harness](https://github.com/deepseek-ai/deepseek-harness) Web UI. The plugin adds a microphone button immediately before the composer send button. It uses the browser's Web Speech API, streams recognition results into the current draft, and never submits the message automatically. ## Features - Adds to the official `conversation.input.right` extension slot. - Supports standard and WebKit-prefixed SpeechRecognition implementations. - Uses the browser language, falling back to `zh-CN`. - Preserves natural spacing for Chinese, Japanese, Korean, and Latin text. - Stops recording when clicked again and aborts when the component unloads. - Requires no additional API key or host-side service. ## Install Install directly from GitHub into the Web profile: ```sh dsh plugin --profile web add github:forrestahha/dsh-voice-input dsh web ``` For a pinned installation: ```sh dsh plugin --profile web add github:forrestahha/dsh-voice-input#v0.1.1 ``` Open the Web UI, select a workspace, and click the microphone button. The browser asks for microphone access on first use. ## Browser support The plugin requires `Speech
Read the full READMERepository license: MIT
dsh-voice-input DSH plugin questions
How do I install the DSH Voice Input plugin?
Run `dsh plugin --profile web add github:forrestahha/dsh-voice-input` in your terminal, then restart the Harness Web UI with `dsh web`. The microphone button will appear in the composer. For a pinned version, use `dsh plugin --profile web add github:forrestahha/dsh-voice-input#v0.1.1`.
Which browsers are supported?
The plugin requires the SpeechRecognition API (standard or webkit-prefixed). Chromium-based browsers (Chrome, Edge, Brave, etc.) offer the best support. If your browser doesn't support it, the microphone button will be disabled but the plugin won't crash.
Do I need an API key or any external service?
No. The plugin uses the browser's built-in Web Speech API. No additional API key or server-side service is required. All processing happens via your browser's speech recognition engine.
Is my audio recorded or sent anywhere?
The plugin itself does not store audio or make any network calls. However, the browser's SpeechRecognition implementation may send audio to the browser vendor's servers for processing. Review your browser's privacy policy if you plan to record sensitive information.
Why does the microphone button not work?
First, check that you are using a supported browser (Chromium-based). Second, ensure the page is served over HTTPS (or localhost). If the button is disabled, your browser likely lacks SpeechRecognition support. Also verify that you have granted microphone permission to the page.