dsh-voice-input
Offline local speech-to-text for DeepSeek Harness with an in-composer microphone and voice transcription tool.
Install
$ dsh plugin --profile web add github:moazzamak/dsh-voice-inputPlugin 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
- Injects a microphone recording button with a 14-bar live volume level meter into the Web GUI composer
- Performs local on-device speech-to-text processing using faster-whisper on CPU without third-party cloud audio uploads
- Appends transcribed speech directly into the active prompt draft without automatically sending it
- Applies an optional intelligent cleanup pass to remove filler words, false starts, and stuttering via the active LLM route
- Provides a voice_transcribe tool usable in headless, CLI, and SDK profiles for batch audio file transcription
Useful For
- Dictating prompts and complex instructions directly into the DeepSeek Harness Web GUI
- Transcribing local audio files like meeting recordings or voice memos in automated CLI workflows
- Operating in air-gapped or privacy-conscious environments requiring zero cloud audio telemetry
Who It Fits
- DeepSeek Harness Web GUI users who prefer voice dictation over typing
- Developers needing local speech-to-text automation in agent sessions
- Privacy-focused teams operating without external speech API access
Documented Limitations
- Requires Python 3.9+ and a one-time local engine setup step via python/setup.py
- Initial model weights must download from Hugging Face unless manually cached or redirected to a mirror
- Audio recording requires browser microphone permissions in supported Chromium or Firefox browsers
- Temporary audio files staged under the OS temp directory require periodic manual clearing
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, 0 development, 0 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
- moazzamak/dsh-voice-input
- Registry source
- GitHub · dsh-plugin topic
- Registry classification
- Plugin Bundle
- Source checked
- 113269e · 2026-10-10
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
- 0GitHub stars
- Forks
- 0
- Open issues
- 0
- Last commit
- 2026-10-06
- Last release
- 2026-10-06
Related DSH Plugins
Ranked by overlapping capabilities, use cases, plugin type, categories, and DSH profile.
Session lifecycle manager for DeepSeek Harness with search, restore, idle cleanup, and diagnosis.
View plugin →Native SQLite-backed task management and project board bundle with Web UI and Agent workflow tools for DeepSeek Harness.
View plugin →Full office suite integration for DeepSeek Harness supporting spreadsheets, docs, slides, bases, and canvases.
View plugin →Agent skill for DeepSeek Harness that turns ideas, plans, and codebases into interactive architecture diagrams.
View plugin →