dsh-voice-input-plugin
A DeepSeek Harness Web composer voice-control plugin for live speech input and spoken assistant replies.
Install
No clear install command was found in the current repository evidence.
Plugin Overview & Capabilities
AI-assisted organization based on the public repository snapshot. The content must be grounded in source evidence and does not replace compatibility or security verification.
Key Capabilities
- Adds a microphone button to the conversation composer tool row.
- Streams Web Speech API transcription into the draft while tap monitoring is active.
- Supports press-and-hold voice messages that send on release.
- Reads assistant replies in sentence-sized streamed segments while generation continues.
- Uses the host /api/tts Edge TTS capability when available, with speechSynthesis fallback.
- Pauses recognition during reply playback to reduce speaker echo, then resumes monitoring.
- Configures recognition language and interim transcription results.
Useful For
- Dictating chat prompts into DeepSeek Harness Web.
- Keeping continuous voice monitoring active while composing and sending multiple messages.
- Using a push-to-talk interaction for voice-chat messages.
- Listening to generated assistant responses without reading the screen.
Who It Fits
- DeepSeek Harness Web users who prefer voice input.
- Users needing Chinese-language speech recognition configuration by default.
- Developers mounting UI plugins and optional Edge TTS capability in a Harness setup.
Documented Limitations
- Microphone input depends on the browser Web Speech API.
- Firefox microphone input is not supported because Firefox has not shipped SpeechRecognition, according to the repository.
- Higher-quality reply reading requires the separate @deepseek-ai/dsh-tts-edge host capability; otherwise the plugin uses browser speechSynthesis.
- Browser speechSynthesis fallback may be less natural and can occasionally be silent on Chrome after an idle gap.
- Version 0.1.1 or later is required due to an incorrect browser bundle ModuleLoader registration in 0.1.0.
DSH Compatibility
Version-specific runtime evidence collected by DSH Plugin. A missing result means we have not tested that combination yet.
Security Signals
Objective signals discovered from package metadata and source inspection. These are not a guarantee that a plugin is safe.
package.json declares 0 runtime, 8 development, 7 peer, and 0 optional dependencies.
package.json declares DSH bundle metadata.
GitHub reports the repository license as MIT.
A root package.json was captured and can be inspected by the registry.
Public GitHub source metadata is available for this registry snapshot.
Source & Registry Notes
Public provenance, Registry classification, and the latest source check for this entry, kept separate from runtime verification.
- Source repository
- Zhangbo-cn/dsh-voice-input-plugin
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin
- Source checked
- bf9ecf7 · 2026-09-18
This project is independently indexed from public source information. DSH Plugin is not affiliated with DeepSeek or the plugin author. Always check the author repository before installation.
Repository Activity
- GitHub stars
- 5GitHub stars
- Forks
- 1
- Open issues
- 2
- Last commit
- 2026-08-25
- Last release
- No release detected
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Selection-annotation plugin for DSH Web that lets users annotate assistant text and send batch annotations with numbered model replies.
View plugin →VSCode-like modular sidebar and bottom panel workbench for DeepSeek Harness with terminal, editor, git, and plugin extension API.
View plugin →Generative UI plugin for DeepSeek Harness that renders interactive components and charts inline from assistant responses.
View plugin →Full context lifecycle management, visualization dashboard, and /context command for DeepSeek Harness.
View plugin →