dsh-speech-service：为 DSH 官方语音输入增加可拆卸的增强能力
dsh-speech-service 是 vb2250158 维护的 DSH 模型推理类插件，在官方语音输入之上提供可关闭的增强：普通输入框话筒与波形、静音 1000ms 收尾、转写后自动发送、新回复自动朗读，以及 FunASR / faster-whisper 本地 ASR 子进程识别。需先启用官方语音与模型；朗读依赖 dsh-rabiroute-agent 与 Rabi TTS。官方能力覆盖后可逐项关闭并卸载。MIT 许可，当前 v0.2.0，适配 DSH 0.2.1-alpha.1 至 0.3.0 之前。
Read More语音生成报告：用 Agent Skill 承接「口述到成稿」的办公链路
「语音生成报告」是 SkillHub 社区 Skill（@user_543d2acf/baogaozc，v1.0.0，office-efficiency），元数据标明 requires_api_key=false。文章基于 SkillHub 公开页与安装文档，说明其定位、可核对的功能边界，以及通过 skillhub CLI 或对话提示安装到 Cursor、Claude Code、Codex 等客户端 skills 目录的步骤；具体触发词与脚本以安装后的 SKILL.md 为准。
Read MoreDeepSeek Harness Plugin: dsh-dictation-hotkey
The DeepSeek Harness (DSH) plugin `dsh-dictation-hotkey` reuses the official SenseVoice engine via a hotkey to provide a “push-to-talk” feature for the Web composer. It supports hold-to-talk or toggle-mode recording, can automatically send or populate text, and provides diagnostic logs. The plugin depends on the DSH voice input bundle, requires installation, and is activated with the Ctrl+Shift+Space hotkey; configuration is stored locally in the browser. It is designed to improve voice interaction efficiency, though the hotkey is only active within the page.
Read MoreDeepSeek Harness Plugin: Turning the Xiaomi Bluetooth Remote 2 Pro into a Voice Input and Quick Control Console
`zmyisme/dsh-xiaomi-remote-2-pro` is a plugin designed specifically for DeepSeek Harness (DSH) to address input interruptions during prolonged use. It integrates the Xiaomi Bluetooth Remote 2 Pro with DSH, providing voice input and quick control features. Key features include: using the remote's voice key to implement “push-to-talk” and insert text; supporting gestures such as single click, double click, and long press mapped to actions like text insertion and sending messages; and offering preset profiles and a learning mode. Installation requires first pairing the remote and enabling DSH's experimental voice input, and the process must be restarted after installation. The plugin is implemented in pure JavaScript, requiring no compilation or administrator privileges, but some system keys (e.g., volume) may not be interceptable on Windows.
Read MoreDeepSeek Harness: Minimal Voice Input Demo
The DeepSeek Harness Web UI Chinese Voice Input Plugin is designed to improve Chinese input efficiency in the Web interface. It relies on the browser's native speech recognition API. By clicking the microphone button on the right side of the input field, it converts speech into text in real time and appends it to the input field. After recognition, it does not automatically send the message, allowing users to edit it manually. Before use, the plugin must be installed via the terminal and the service must be restarted. Use cases include quick Chinese input and assisted note-taking. Note that it only supports the Web platform and the Chrome browser, and audio processing relies on browser services, so offline availability is not guaranteed.
Read Moredsh-voice-input-plugin: DeepSeek Harness Web Voice Input Plugin
`dsh-voice-input-plugin` is a zero-dependency plugin for DeepSeek Harness that uses the browser-native Web Speech API to perform local speech-to-text. The plugin requires no build step, provides a microphone button through the Web GUI toolbar, supports real-time word-by-word transcription and continued editing while speaking, and does not upload audio to the backend. Installation requires DSH version `>=0.1.0-rc.5`, and the Web Profile must be restarted for it to take effect. It currently only supports Chrome and Edge browsers, relies on an online speech recognition service, and the recognition language follows the browser's language settings.
Read MoreIntroducing the DeepSeek Harness Voice Input Plugin: Local Multilingual Speech Recognition and Codex Desktop Dictation
The DeepSeek Harness (DSH) plugin `dsh-dictation` provides local multilingual speech-to-text capabilities and integrates with Codex Desktop to write into drafts. It follows an “everything-is-a-plugin” architecture and is suitable for local voice input scenarios. Due to a failed README fetch, specific installation steps and operational commands are unavailable; consulting the GitHub repository directly is recommended. When installing, note that the plugin runs with the permissions of the current DSH process, so the source code and license must be reviewed to ensure security and compliance.
Read MoreDeepSeek Harness Voice Input Plugin: Using the Microphone in Composer
The DeepSeek Harness (DSH) plugin `yafangwang9/dsh-voice-plugin` provides microphone input capability for Composer components and supports speech-to-text conversion. Users need to run the installation command and restart the process, then select a speech recognition model in the settings (Doubao Voice for Chinese and OpenAI for other languages) and enter the API key, after which they can perform voice input through the microphone in the interface. As a community-maintained project, the plugin requires careful review of its source code and license before installation.
Read MoreDeepSeek Harness Voice Input and Text Cleanup Plugin
The dsh web voice input plugin provides users with voice-to-text and AI text-optimization features. Its core functionality includes holding Right Alt to speak and record, with support for browser-local recognition and multiple cloud recognition APIs (such as OpenAI Whisper). After recording, users can left-click to correct the text or right-click to correct and format it; the result replaces the original text in place. The plugin stores configuration settings and API keys securely in the local browser. It must be installed using the official command and dsh web restarted before use.
Read Moredsh-mic-dictation: DeepSeek Harness Voice Input Plugin
dsh-mic-dictation is a native client plugin for DeepSeek Harness, designed to address the tedium of manually entering tasks on the web. By inserting a microphone button into the prompt bar, it leverages the browser’s native speech recognition capability to automatically convert speech to text input, eliminating the need to switch input methods. The plugin runs entirely client-side, works out of the box after installation, does not collect any data, and supports Chrome, Edge, and Safari. After clicking the button and granting microphone permission, users can simply speak their instructions, which are then automatically filled into the prompt box.
Read Moredsh-voice-input: DeepSeek Harness Voice Input Plugin
`dsh-voice-input` is a voice input plugin for DeepSeek Harness that converts speech to text using browser native APIs or an OpenAI-compatible `/audio/transcriptions` endpoint. The plugin integrates a microphone button into the input field, supports real-time status feedback, engine selection (browser, cloud, or local), and secure credential management. It is zero-dependency, requires no build steps, and provides one-click install and uninstall scripts, with support for Whisper local services and Ollama (models with audio projection capabilities).
Read Moredsh-vox-input: Serverless DSH Web Voice Input
dsh-vox-input is a client-side plugin that provides speech-to-text functionality for the DSH Web composer. Built on the browser's native Web Speech API, it requires no server, API key, or model download, ensuring local data processing. The plugin supports real-time recognition and switching between Chinese and English, and features visual waveform control, real-time caption streaming, and intelligent backfill. It is suitable for privacy-focused use cases and for reducing development costs. Installation requires creating a symbolic link, configuring cordis.patch.yml, and restarting the service.
Read Moredshtools-sensevoice-input: A Local Voice Input Plugin Based on SenseVoiceSmall
The DSH Desktop plugin `dshtools-sensevoice-input` provides a fully local voice interaction solution that does not rely on cloud APIs. By leveraging the FunASR framework and the SenseVoiceSmall model, the plugin integrates a microphone button; when clicked, recorded audio is automatically transcribed and written to the draft area via `inputActions.setDraft`, along with language and emotion labels. It supports theme adaptation, lazy model loading, and self-healing capabilities. Before use, a script must be run to install the model (approximately 5GB). It currently operates in offline batch-processing mode, with a maximum recording duration of 60 seconds per session. Initial loading may take longer, but subsequent recognition is fast.
Read MoreIntroduction to the DeepSeek Harness Voice Input Plugin
`dsh-voice-input` is a voice input plugin for DeepSeek Harness. It uses the browser's native Web Speech API to provide real-time speech-to-text without requiring a self-hosted service. After installation, users can click the microphone button on the left side of the input box to start or stop recording. The recognized text is appended to the input draft in real time and can coexist with manually typed content. The plugin is compatible with the DSH theme, but it is not available in Firefox and requires an internet connection to use cloud-based recognition services.
Read Moredsh-voice-scribe: Dedicated Voice Input Plugin for DSH
`dsh-voice-scribe` is a third-party DSH plugin maintained by PensiveFei, designed to provide a complete and robust local offline speech recognition experience. **Core features** include: support for the local SenseVoice engine (with automatic fallback to browser or cloud services), custom hotkeys and a push-to-talk mode, streaming real-time feedback, microphone diagnostics, and multilingual support. Advanced features include a hotword replacement table, AI polishing (reusing DSH models), and more. **Installation and usage**: it can be installed via the command line and is compatible with both Electron and Web clients. Use the default Alt hotkey for recording, and customize key combinations as needed. If recognition issues occur, check the microphone device settings. The plugin has limitations when the input contains references or under mainland China network conditions, but it offers richer functionality than the official plugin and places a stronger emphasis on privacy.
Read MoreDeepSeek Harness Voice Input Plugin (Available in China)
`dsh-voice-input-cn` is a voice input plugin developed specifically for DeepSeek Harness (DSH) and available in China, maintained by Schumchanvi. It relays requests through a local Python bridge, resolving the issue that the original version relies on Google APIs and is inaccessible in China, and implements speech-to-text based on Alibaba Cloud DashScope ASR. Core features include clicking the microphone to start speaking, real-time text input, support for switching between Chinese and English, and automatic submission after recognition. Installation requires configuring a Python 3.10+ environment with necessary dependencies, starting the local bridge service, and entering an Alibaba Cloud API key in the browser. The plugin is secure and reliable, listens only on the local loopback address, sends audio only to Alibaba Cloud for recognition, and is suitable for use in China.
Read Moredsh-voice: A Local Voice Input Plugin for DeepSeek Harness
dsh-voice is a DeepSeek Harness plugin that provides local speech-to-text (STT) for the Web UI and TUI. It uses the host-side ONNX-based Whisper service and integrates the microphone in the browser, enabling privacy-safe voice input without external binaries. To install, add it to a DSH Profile; configuration is optional. It supports downloading various Whisper models from Hugging Face. Note that TUI mode depends on system recording tools, while in Web mode audio is recorded locally and transferred over HTTP.
Read Moredsh-voice-input-space: A Zero-Dependency DeepSeek Harness Voice Input Plugin
dsh-voice-input-space is an open-source voice input plugin in the DeepSeek Harness (DSH) ecosystem, maintained by developer XSakura666. Released under the MIT license, it does not rely on local models and instead uses the browser's native Web Speech API to provide minimal speech-to-text via a long-press of the spacebar. The plugin supports Chinese (Mandarin, Traditional Chinese, Cantonese), English, Japanese, and Korean, and features Chinese input method compatibility, status feedback, and instant cancellation. After installation, users simply need to long-press the spacebar in any input field to start listening; releasing it inserts the transcription into the field. The plugin requires no model downloads, but needs an internet connection to access the speech services.
Read MoreParakeet Voice Input for DeepSeek Harness: A Local Offline Voice Input Plugin
Parakeet Voice Input is a local speech-to-text plugin developed by vTRKA for DeepSeek Harness. By integrating the Parakeet and NeMo-Speech.cpp models, it allows users to record audio in Composer using a keyboard shortcut (Ctrl+E by default) or a button. All audio is processed locally and is not uploaded to external APIs. Installation requires running a script in PowerShell and restarting the application, and users must ensure that local port 8080 is not in use. This solution is suitable for developers who value privacy and offline operation.
Read MoreIntroduction to the DeepSeek Harness Voice Input Plugin
`dsh-voice-input` is a local speech-to-text plugin for DeepSeek Harness. Built on FunAudioLLM's SenseVoice model, it integrates microphone support into the Web Profile to automatically transcribe recordings. It does not upload audio to the cloud, helping protect privacy. Installation requires Python 3.10+ and Node.js >= 22.19.0. The plugin supports one-click recording, automatic model download management, and background service startup. The first use requires waiting for the model download, and the project is released under the MIT open-source license.
Read Moredsh-plugin-voice-input: DeepSeek Harness Voice Input Plugin
The voice input plugin for DeepSeek Harness (DSH) aims to address the typing efficiency bottleneck on the web. Maintained by NewDaNew, the plugin injects a microphone button to enable speech-to-text conversion. Its core features include local speech recognition (without uploading data), automatic sending, and language configuration, with real-time transcription displayed in the input box. Users can install it with one click via the official script, but must ensure they are using Chrome/Edge and DSH version 0.1.0-rc.6+. The plugin is released under the MIT open-source license.
Read More@motong/dsh-voice: Voice Input and Text-to-Speech Plugin for DeepSeek Harness
`@motong/dsh-voice` is a community plugin that provides voice capabilities for DeepSeek Harness (DSH). It requires no external API key and supports speech-to-text (STT, with hotkey invocation) and text-to-speech (TTS, based on Microsoft Edge voices). Users can dictate using the `Ctrl+Shift+Space` shortcut or by clicking a button, and switch between voices in settings. The plugin relies on native browser features and is compatible with Chrome/Edge; TTS requires an internet connection. It is open-sourced under the MIT license.
Read Moredsh-voice-input: System-Level Voice Input
`dsh-voice-input` is a plugin for DeepSeek Harness, designed to enable global voice transcription in non-plugin environments. It leverages a local Whisper model to transcribe speech into text in real time via hotkeys and paste it at the cursor position. Installation requires running DSH with administrator privileges. Main features include holding F2 for voice transcription, pressing F3 to record audio only, and managing the history through the system tray. The plugin supports configuring the recognition mode, language, and device, and requires downloading approximately 2.5GB of dependencies and models upon first use.
Read MoreDeepSeek Harness Voice Dialog Plugin Introduction
The DeepSeek Harness plugin `lak321/dsh-voice-chat` provides fully client-side voice interaction without requiring server-side support. Its core features include using the Web Speech API for speech input and text-to-speech (TTS) output, with support for multiple languages and adjustable speech rate and pitch; automatically removing Markdown markers to prevent unwanted noise; and saving settings via localStorage. It can be installed with `dsh plugin --profile web add "github:lak321/dsh-voice-chat#<commit>&path:/"`. Development requires adherence to React Hook rules and the `exports.inject` export requirement.
Read Moredsh-voice-input: DSH Local Voice Input Plugin
The DeepSeek Harness voice input plugin, `dsh-voice-input`, is designed to provide a local, streamlined, and precise speech-to-text experience. It integrates a local Whisper engine, enforces output in Simplified Chinese, and features hallucination prevention, privacy protection, optimization for users in China, and high performance. On first use after installation, it automatically downloads a model of approximately 500 MB. It supports click-to-record transcription via the microphone icon and also allows switching to a cloud API. It is suitable for users who find typing inconvenient or who prioritize privacy.
Read Moredsh-voice: DeepSeek Harness Voice Input Plugin
dsh-voice is a voice input plugin for DeepSeek Harness (DSH). It adds a microphone button to the bottom-right corner of the input box and supports one-click recording-to-text. The plugin provides three speech recognition engines: `webspeech`, `xfyun` (iFlytek), and `whisper`, and supports multi-device selection, automatic gain, and stitching of long utterances. Installation requires executing a specific command and restarting the service, with the configuration file located in `client.js`. Using the iFlytek engine requires setting local environment variables or the configuration file. On Chrome in mainland China, `webspeech` may cause errors, and the recognition language for `whisper` must be set manually.
Read Moredsh-voice-input: Chinese-English Voice Input Plugin for DeepSeek Harness Web
DSH (DeepSeek Harness) implements web-based voice input through the `dsh-voice-input` plugin. The plugin is built on the browser-native Web Speech API, requires no server-side support or API key, and supports speech recognition for Chinese, English, and mixed-language speech. Users can click the microphone icon to the left of the web input field to start speaking, and the recognized text is automatically inserted into the input field. This plugin is suitable for quickly capturing ideas or entering conversational instructions, but recognition accuracy depends on the browser’s speech service, and it must be used in a browser that supports the Web Speech API, such as Chrome or Edge.
Read Moredsh-speech-input: Add a Microphone Button to the DeepSeek Harness Web Input Field
# Preface In dsh web, conversing with a model often requires typing out long prompts for several minutes. Since the input box lacks a voice input entry point, dictation usually relies on system dictation and pasting, or directly modifying the Harness DOM to inject controls—both of which are unreliable, with the latter specifically being prone to breaking after page updates. The `dsh-speech-input` introduced below takes a different approach: it adds a native microphone button to the right of the input box using official slots. Clicking it starts dictation, writing recognition results to the current draft in real-time. Clicking it again stops the process and appends punctuation to the end of the sentence, ensuring the message is never sent automatically. DSH's philosophy is "Everything is a plugin," and input box enhancement falls squarely within the scope of the plugin system. # What is this `dsh-speech-input` is lizne
Read Moredsh-web-speech-input: Add a microphone button to the DeepSeek Harness Web input field
dsh-web-speech-input is a speech-to-text plugin for the DeepSeek Harness Web UI (maintained by leozou320-ai, v1.0.0, MIT license). Its core functionality is adding a microphone button to the left side of the prompt composer. Upon clicking, it uses the browser's Web Speech API to write real-time speech recognition results into the current draft. It supports interim and final recognition text, automatic language selection (Page Language -> Browser Language -> zh-CN fallback), clear status indicators, and accessibility support. The plugin never sends a prompt automatically; the submission action is always performed by the user. In terms of privacy, microphone permission is only requested after the user clicks. There are no server-side components, no audio storage, and no network requests are sent; audio processing depends on browser implementation. Installation Requirements
Read Moredsh-dictate: Integrate voice input into DSH Composer
dsh-dictate is a context-aware voice input plugin designed for DeepSeek Harness (DSH) Composer. It facilitates voice input for long prompts or code explanations, preserving transcriptions in the input box for user review and modification, and does not initiate a bidirectional voice session. By default, the plugin uses the Web Speech API in Chrome or Edge for browser-based recognition, displaying real-time transcriptions during recording. On Apple Silicon Mac and Windows x64 platforms, users can choose to employ the local SenseVoice model for enhanced transcription, requiring the download of approximately 253 MB of model files. Its core advantages include context-awareness (optimizing wording based on the last 6 conversations), optimized recognition for Chinese-English mixed text, and strict send controls (defaulting to no automatic sending).
Read Moredsh-better-input: Add voice input, AI polish, and file-to-Markdown for DeepSeek Harness
dsh-better-input is an input enhancement plugin for the DeepSeek Harness (DSH) Web UI, maintained by DIAG5 and licensed under the MIT License. The plugin aims to optimize the input paths for long prompts, local documents, and dictated content. Its main features include: 1. **Voice Input:** Local recognition based on the Web Speech API, requiring no API Key, featuring real-time streaming transcription, and supporting custom recording durations. 2. **AI Polishing and Optimization:** Reuses the models configured in DSH to automatically clean up speech recognition errors, remove filler words, and precisely optimize prompts, supporting original text comparison and undo operations. 3. **Prompt Templates:** Templates are stored locally; typing `/` allows for quick insertion, ensuring data privacy. 4. **File Handling**
Read MoreDSH Web Voice Input Plugin: anweat/dsh-voice-webspeech
`anweat/dsh-voice-webspeech` is a browser-based speech input plugin for the DSH Web GUI, released under the MIT license. Leveraging the browser's built-in Web Speech API, this plugin implements a "press and hold to speak, release to transcribe" function. It converts speech to text in real-time and inserts it into the input box, supporting subsequent editing or configuration for auto-sending upon release. Its core advantages are zero server-side processing, zero API keys, zero model downloads, and zero Python dependencies. Audio is processed solely by the browser; it does not pass through the DSH server, is not saved to disk, and contains no telemetry, ensuring privacy and security. The plugin supports multiple BCP-47 languages, including Chinese, English, Japanese, and Korean, with preferences stored locally in the browser. Regarding environment requirements, it requires DSH version `>=0.1.0-rc.3 <0.2.0`, and N...
Read Moredsh-audio-copilot: Add voice input to DeepSeek Harness chat box
dsh-audio-copilot is a DSH voice input plugin maintained by ai-yucheng, based on DSH 0.1.0-rc.7 and licensed under the MIT license. This plugin allows users to click the microphone button in the chat box to speak, automatically filling the transcribed text into the input box via multi-engine ASR. It supports four engines: zhipu, local, gemini, and openai. zhipu excels at Chinese dialects and foreign languages and operates on a pay-as-you-go basis; local uses faster-whisper for offline transcription and requires downloading a model of approximately 460MB to 1.5GB; gemini's free tier includes rate limiting and an automatic retry mechanism; and openai is available as an optional engine. Prerequisites include DSH Desktop, Node.js ≥ 20, and the PATH environment variable.
Read Moredsh-voice-kit: Add voice input and audio response for DeepSeek Harness Web GUI
`dsh-voice-kit` is an MIT-licensed voice plugin for the DeepSeek Harness Web GUI, designed to enhance the interaction experience by primarily addressing the needs of voice input and reading aloud responses. The plugin supports voice input via a microphone, appending transcription results to the draft. It also supports three types of text-to-speech engines: System, Microsoft Neural, and Cloud Neural (such as SiliconFlow CosyVoice2), capable of automatically cleaning up Markdown and streaming long text in chunks. Users can configure the engine, speed, pitch, and recognition language in the settings; the key status only displays the saved state without revealing the actual values. Installation requires executing `dsh plugin add dsh-voice-kit`, followed by restarting DSH Desktop and refreshing the GUI. The plugin requires...
Read Moredsh-smart-input: Voice Input and Prompt Optimization Plugin for DeepSeek Harness
dsh-smart-input is a DeepSeek Harness (DSH) Web client plugin maintained by v-quest123456 under the MIT license, designed to address the issues of incomplete manual prompt writing and the lack of feedback in voice-to-text. Its core features include voice input and prompt optimization. Voice input supports the browser Web Speech API, local Whisper ONOX, and the online Whisper API, featuring real-time streaming transcription, smart punctuation, multi-language support for Chinese, English, and Japanese, and a continuous conversation mode, with the shortcut key being Ctrl/Cmd + Shift + V. Prompt optimization offers six strategies, including automatic, simplification, and structuring, supporting context-awareness (injecting workspace file information), before-and-after comparison, and one-click acceptance, with the shortcut key being Ctrl/Cm.
Read Moredsh-plugin-voice-input: Local voice to draft in DSH Web input
`dsh-plugin-voice-input` is a DeepSeek Harness Web voice input plugin developed by tangbut1, open-sourced under the MIT license. This plugin aims to address the inconvenience of inputting long sentences or mixed Chinese and English content in web input boxes. By adding a "hold to speak" microphone button next to the input box, it enables local offline speech recognition. Users press and hold the button to record audio, and upon release, the system performs offline recognition using the `sherpa-onnx-node` and `SenseVoice-SSmall INT8` models via the local loopback address (127.0.0.1:18765). It supports automatic language detection for multiple languages, including Chinese, English, Japanese, Korean, and Cantonese. The recognized results are appended to the current draft via DSH's official slots and events, and are not sent automatically.
Read Morevoco-input-sh: VocoType Voice Input Plugin for DeepSeek Harness
voco-input-sh is a DSH web plugin designed to integrate VocoType's local offline speech recognition into the chat input box, reducing reliance on external tools. Licensed under the MIT License (version 0.2.0), its main features include providing a microphone button next to the input box to trigger VocoType for speech recognition, automatically inserting results into the draft area, and featuring intelligent deduplication. It supports Paraformer Chinese recognition, Voice Activity Detection (VAD), and punctuation prediction, and is capable of automatically detecting, launching, and even silently installing VocoType (requiring the download of approximately 25MB for the installation package and 1.6GB for the models). Installation requires adding a local package in the DSH profile directory using pnpm, configuring the mount in `cordis.patch.yml`, and then restarting.
Read Moredsh-stt-input: Add speech input for DSH Web GUI
`dsh-stt-input` is a voice input plugin for the DeepSeek Harness (DSH) Web GUI, designed to address the need for quickly inputting long prompts via voice within DSH. The plugin allows users to record by clicking the microphone button and fills the recognized text directly into the input box upon stopping. It provides two recognition engines: one is browser-based local recognition using the Web Speech API, supporting zero-configuration usage in Chrome/Edge with real-time intermediate results; the other is API recognition based on MediaRecorder and OpenAI-compatible interfaces, supporting OpenAI, Groq, or custom services and multiple Whisper models. The plugin supports configuring service presets, API keys, language, and write mode (append or replace). API
Read Moredsh-mic-input: DSH Web UI Browser Microphone Input Plugin
`dsh-mic-input` is a DSH Web UI speech input plugin maintained by QT-Chen and released under the MIT license. It runs entirely on the client side, requiring no server-side processing, API keys, or model downloads, utilizing the Web Speech API for microphone dictation. Core features include adding a microphone button to the input toolbar to initiate real-time dictation; preventing text duplication through a draft reconstruction and deduplication mechanism; supporting up to six silent continuations to maintain the session; providing intelligent punctuation (auto-completion, preservation, and removal) and language selection for Chinese, English, or following the system language; and supporting optional automatic message sending. Users can install it from GitHub or npm, requiring a Node.js version of 20.0.0 or higher. After installation, restart DSH Web or refresh the page for the changes to take effect.
Read Moredsh-voice-input-web: Adding Speech-to-Text Transcription to DeepSeek Harness Web Input Fields
dsh-voice-input-web is a client-side plugin for the DeepSeek Harness Web UI, designed to implement speech-to-text functionality through the browser's Web Speech API, requiring no API key or server-side modifications. The plugin adds a microphone button to the chat input toolbar; users can click it to start or stop listening, with real-time recognition results displayed in a floating overlay and the final text automatically appended to the input draft. It supports Chinese, English, and automatic language switching, with an optional auto-send feature. Installation methods include online and offline modes: for online installation, simply execute the `dsh plugin --profile web add dsh-voice-input-web` command and restart the service; for offline installation, manually place the package file and configure the patch file.
Read More@zhangbo-cn/dsh-client-ui-voice-input: Adding Composer Voice Input and Reply Read-Aloud to DeepSeek Harness Web
`@zhangbo-cn/dsh-client-ui-voice-input` is an MIT-licensed plugin maintained by Zhangbo-cn, designed to enhance voice interaction for the DeepSeek Harness Web UI. This plugin adds a mic button in the composer area and implements speech-to-text based on the browser's Web Speech API, requiring no API key. Core features include: clicking the microphone starts listening and writes the recognition results into the draft in real time, character by character; holding the microphone, speaking, and releasing automatically sends the message; when conditions are met, it reads the assistant's reply aloud via the host's Edge TTS or the browser's `speechSynthesis`. The plugin supports silently restarting recognition to maintain continuous listening and offers optional automatic sending configuration. Installation only requires
Read Moredsh-ears: Open Source Voice Input Plugin for DeepSeek Harness
`dsh-ears` is an open-source voice input plugin for DeepSeek Harness (DSH) (v0.2.1, MIT license), designed to enhance the text input experience in the Web UI through speech recognition and optional text polishing. This plugin supports multiple recognition backends, including Web Speech, local Whisper, Groq, Deepgram, Alibaba Cloud Bailian, Tencent Cloud, Xiaomi MiMo, SiliconFlow, and custom OpenAI-compatible interfaces. Its core workflow converts voice input into an editable draft, with the polishing feature enabled by default to refine text, but it does not automatically send messages, ensuring users retain full control. Installation requires DSH version 0.1.0-rc.6 to 0.1.1-rc.2 and Node.js ^22.19.0 or >=24.
Read MoreAdd Voice Input and Speech Synthesis Functions to DeepSeek Harness with dsh-voice
DSH-voice is a plugin for the DeepSeek Harness tool maintained by Jesse-njx. It uses `transcribe` to turn dictation or audio files into user messages, uses `speak` to read content in the background without blocking the conversation turn, and uses `/voice` to toggle automatic reading of replies per session. Audio files are saved to `~/.dsh/voice/` by default, and only references and transcriptions are saved in session logs. It is licensed under MIT, with version 0.1.0. For installation, please refer to the directory command: `dsh plugin add github:Jesse-njx/dsh-voice`. By default, it uses an offline backend, and OpenAI and Edge TTS will not be automatically selected; v0.1 does not support real-time intercom or wake words.
Read More