dsh-voice-notify: DSH Plugin: Chinese Voice Notification at the End of Each Conversation Turn
`dsh-voice-notify` is a Windows plugin for DeepSeek Harness (DSH) that hooks native extension points and plays a Chinese voice notification at the end of a conversation. It includes 20 built-in “Xiaoxiao”-style reduplicated phrases and supports visual configuration of the global toggle, sub-agent filtering, and minimum interval. Installation requires only a single PowerShell command and supports daemonized hot mounting. Its non-blocking playback mechanism ensures that task finalization is not affected, and it supports custom voice libraries and scripts.
Read MoreDeepSeek Harness Voice Control Plugin: Voice Input and Speech Output
The `dsh-voice-control` plugin leverages the Web Speech API to add voice interaction capabilities to the DeepSeek Harness Web interface. It supports voice input via the microphone icon or the `Ctrl+M` shortcut (with auto-send or draft insertion) and automatically reads aloud the assistant’s responses (up to 1,200 characters). The plugin includes a right-click menu for adjusting speech rate, selecting a voice, and previewing playback, and it runs entirely locally. Note that it only supports Chrome/Edge browsers, and speech recognition requires network access.
Read MoreDeepSeek Harness Voice Playback Plugin: Automatic Reminders and Sound Effects Library
The DeepSeek Harness (DSH) Voice Broadcast plugin provides developers with auditory feedback, addressing the lack of audio prompts in CLI tools. It automatically announces the end of a conversation or errors, supports 20 zero-configuration built-in sounds, and offers vol cano's “Voice Cloning” for timbre cloning. Compatible with Bluetooth headsets, it does not interrupt music and features millisecond-level logging to ensure intuitive and effective status feedback.
Read Moredsh-tts: Voice broadcast plugin for dsh web final results
dsh-tts is a dsh web text-to-speech (TTS) plugin developed by vim0x3c (version 0.1.0, MIT license), designed to reduce the reading cost of long text through voice synthesis. Utilizing Edge TTS technology, this plugin reads only the final assistant response while ignoring intermediate steps. It supports segmented serial reading of long replies, speed adjustment ranging from 0.5x to 2x, and offers over 20 voice options. It provides built-in approval sounds or custom audio functionality and automatically falls back to the browser's local Web Speech API when Edge cloud synthesis fails. The plugin has no third-party dependencies, supports offline installation, and requires installation into the web profile via a local tarball combined with the `-w` flag. After installation, the local bridge service binds to 127.0.0.1 and occupies ports 8765-8780 to provide synthesis and health check interfaces, setting...
Read Moredsh-plugin-notify: Notification Egress Plugin for DSH
dsh-plugin-notify is a DeepSeek Harness plugin designed for the Windows environment, aimed at addressing the issue where users cannot receive timely updates on the Agent's status when away from the computer. It enables the Agent to invoke the `notify_user` tool by injecting system prompts, supporting proactive notifications upon completion of long-running tasks, errors, or when confirmation is required. The plugin provides an automatic fallback mechanism, including automatic alerts on session errors, desktop notifications at the end of turns, and Chinese voice calls when there is no interaction within a timeout period. Notification channels cover desktop pop-ups (toast), Chinese speech announcements (speak), and system alert sounds (sound), and support four synthetic sound effects such as "explosive" and "victory." Users can manually send notifications via the `/notify` command and adjust settings through environment variables or the configuration panel.
Read Moredsh-speak: Adding Voice Broadcast to DSH and Other AI Coding Harnesses
dsh-speak is a DSH web plugin maintained by Alan2Z, designed to address the pain point of users missing screen prompts during long tasks. It utilizes system speech synthesis technology to read aloud in real-time the final responses, approval requests, and agent queries from DSH sessions. The plugin supports both Windows (SAPI5) and macOS (say command), prioritizes natural speech, and includes automatic cleaning of Markdown, URLs, and emojis to adapt to speech synthesis. Its core features include: monitoring event streams and broadcasting key information, a best-effort operating mode (non-blocking, no exceptions thrown), and the Bundle auto-registration feature introduced in version 1.3.0, which simplifies installation and configuration. The installation is flexible, supporting npm or gi...
Read More