In-depth architectural comparison of the Whisper Windows MCP and QuillHub MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.
At a Glance & Executive Verdict
Whisper Windows MCP
Speech-to-Text · Local stdio
Quality: 64/100 (Good) | Auth: No auth required
QuillHub
Speech-to-Text · Remote HTTP/SSE
Quality: 52/100 (Good) | Auth: No auth required
Verdict Summary: Choose Whisper Windows MCP if you need specialized Speech-to-Text tools running via a local process. Choose QuillHub if your workspace requires Speech-to-Text integration with remote web transport. Both servers can be configured concurrently in your client's mcpServers manifest.
Which MCP Server Should You Choose?
Choose Whisper Windows MCP when:
You need dedicated capabilities in the Speech-to-Text domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
You have access to required keys: WHISPER_CLI_PATH, WHISPER_MODEL.
Windows-native local audio and video transcription using whisper.cpp with Vulkan GPU acceleration. No cloud APIs, no Python. Batch processing, multilingual support, model management, and background job handling built in.
Meeting transcripts for AI agents: search calls, read who said what, transcribe files and links.
Transcribe a single file. Supports blocking (default) or background mode for long files.
check_progress
Monitor a background transcription job started with `transcribe_audio` (background=true).
start_batch
Automated sequential batch transcription of all untranscribed files in a folder. Sorts by duration (shortest first), processes one at a time as background jobs, validates each output. Batch self-advances when each file finishes — no polling required.
check_batch_progress
Monitor a running batch. Automatically advances to the next file when the current one finishes. Returns overall progress, current file with timestamp, and any failed files.
generate_subtitles
Generate subtitle files. Supports automatic language detection and English translation output. Outputs SRT (widest compatibility) or WebVTT (web and HTML5 video).
Ready-to-Paste Client Configurations
Paste either (or both) of these JSON server blocks into your client config file (e.g. claude_desktop_config.json or ~/.cursor/mcp.json).
Whisper Windows MCP is categorized under Speech-to-Text and uses a local stdio subprocess. In contrast, QuillHub belongs to Speech-to-Text using remote streaming HTTP/SSE transport. Select Whisper Windows MCP when you need capabilities focused on speech-to-text and QuillHub when you require tools for speech-to-text.
Analyze files before committing to transcription. Returns duration, size, codec, and estimated transcription time on CPU and GPU. For folders, shows all files in a sortable table with transcription status.
check_config
Verify whisper-cli.exe, the model file, and FFmpeg are all accessible. Run this first if anything is failing.
list_models
List all Whisper model files installed in your models directory. Shows filename, size, whether it is currently active, quantization status, and recommended use case. No network calls — reads local filesystem only.
download_model
Download a Whisper model directly from Hugging Face into your models directory. Only downloads from trusted Hugging Face namespaces. After downloading, use `switch_model` to activate it.
switch_model
Switch the active Whisper model for the current session without restarting Claude Desktop. Change is session-scoped — does not persist after restart. To make permanent, update `WHISPER_MODEL` in your config.
check_system
Detect GPU hardware and verify Vulkan acceleration is available. Reports GPU name, VRAM, whether `ggml-vulkan.dll` is present, and recommends the best model size for your hardware.
whisper_server
Start, stop, or check the **persistent model server** (whisper.cpp's `whisper-server`). While running, the active model stays resident in VRAM and every `transcribe_audio` / `transcribe_batch` call is served over localhost with **no per-file model reload** — a large speedup when transcribing many s…
QuillHub Tools (16)
search_meeting_transcripts
Full-text search across meetings in a workspace: titles, summaries, decisions, action items and transcript text. Morphological matching in English and Russian.
list_transcriptions
Paginated list of recordings with summary, key theses, decisions, action items, participants, project. Filter by text, date range, workspace, project.
get_transcription
One recording in the format you need: `summary`, `text`, `segments`, `dialog` (who said what), `paragraphs`, `chapters`, `subtitles`. Supports time windows for long calls.
get_person_speech
Timestamped quotes from one person across every meeting, with filters by date and project.
find_subjects
Find tracked subjects (people, deals, projects, clients, candidates, custom types) with their AI-maintained note, open commitments and mention counts.
get_subject_page
Full page for one subject: living note, open commitments, every meeting mention with quotes and speakers, unresolved questions.
get_state_timeline
How a subject's note changed over time: revisions with diffs and the recording that caused each change.
list_open_questions
Contradictions and ambiguities the AI found across meetings (e.g. two different budget figures) that need a human decision.
get_project_memory
AI-maintained briefing for a project or workspace: purpose, glossary, key people, activity digest.
list_subject_types
Subject types available in a project and the fields each one tracks.
list_workspaces
Your workspaces and the projects inside them, so the agent can read teammates' shared recordings.
create_transcription
Transcribe a URL (YouTube, direct file link) or an inline base64 file up to 25 MB.