MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Self-hosted, multi-user transcription platform: record or upload audio. Speaker-labeled, timestamped transcripts, Recognize speakers across recordings, Summarize, extract action items and chat over your transcripts with your own OpenAI-compatible LLM. Your Audio, your Server, your Model. Tested: Laptop RTX4070, RTX3090, RTX5090 and ROCM Strix Halo
npm AGPL-3.0 ★ 9 -
Local-first Audio Intelligence MCP server: transcription, diarization, sound events, quality, timelines, grounded Q&A, search
pypi Apache-2.0 ★ 2 -
MCP server for the Orchard speech-to-text API: transcription and speaker diarization for real-world Spanish and Portuguese audio.
MIT -
Self-hosted speech-to-text: GPU transcription API (GigaAM, Whisper, pyannote diarization) + web portal, queue, MCP connector
Apache-2.0 -
MCP server that turns meeting audio into speaker-labeled transcripts — chunked Whisper transcription, diarization, and alignment, returned as structured data. Built on FastMCP.
-
Thin stdio MCP connector for the Pepys transcription API. npx pepys-mcp – give Claude, ChatGPT, or Cursor audio/video transcription: diarization, SRT/VTT, podcasts, paste-a-link.
MIT -
@speakagent/mcpgithub.com/legolev/speakagentupdated this year
Model Context Protocol server for SpeakAgent — transcribe, diarize, build meeting protocols and extract todos from audio/video in Claude Desktop, Cursor, VS Code and other MCP-aware AI clients.
npm MIT -
stdio MCP server for the Meeting Transcriber macOS app: submit audio, read diarized transcripts, and control meeting watching over its localhost automation API. Fully on-device.
MIT -
Local-first audio and speech dataset curation studio for waveform review, transcription, diarization, consent, quality, segmentation, exports, MCP, and modular AI plugins.
MIT ★ 2 -
Local offline speech recognition for Russian speech with German words: ASR model ensemble with word-level voting, diarization, desktop app, CLI and MCP server. Research notes included.
GPL-3.0 -
Windows desktop app for cloud audio transcription and speaker diarization (AssemblyAI / Deepgram / Gladia / Speechmatics) with KZ+RU+EN code-switching, LLM task extraction to Linear / Trello / Glide, and meeting protocol generation
MIT ★ 1 -
Local RAG archive of your calls: faster-whisper transcription, pyannote diarization, entity graph and tasks, hybrid vector search with an "ask the archive" chat, live call co-pilot, Telegram and document ingestion, MCP server for Claude. Runs on your machine.
MIT ★ 2 -
ListenMiao: local web app for MediaTek Research's Breeze ASR 25/26. Turn Taiwanese Mandarin, Mandarin–English code-switching and Taiwanese Hokkien into subtitles and transcripts without uploading audio. Hotwords, speaker diarization, subtitle editor, SRT/DOCX export, burn-in, CLI, MCP server, Agent Skill; CPU, Intel GPU or NVIDIA/AMD GPU.
MIT ★ 2 -
MCP server for podcast transcription with speaker diarization, optimized for Apple Silicon
-
dicoclerkgithub.com/JinwangMok/dicoclerkupdated this year
Discord voice channel meeting clerk — real-time STT with speaker diarization and meeting minutes generation. MCP server for Openclaw integration.
MIT -
plaud-mcpgithub.com/Namraks-Labs/plaud-mcpupdated this year
Local MCP server that syncs Plaud Note Pro recordings (AI summaries + diarized transcripts) into markdown files.
MIT ★ 3 -
freeaudiototext-mcpgithub.com/double2dev/freeaudiototext-mcpupdated this year
An official MCP Server for FreeAudioToText.com - Turn any audio/video into text 100% free with speaker diarization.
-
Offline speech-to-text & speaker diarization for AI agents — Whisper ASR + MCP server + CLI + Python bindings, fully on-device, no cloud | 离线语音识别与说话人分离,Agent 开箱即用
pypi MIT ★ 3 -
Self-hosted AI knowledge base with hybrid semantic search (pgvector + FTS + RRF), MCP server, multi-provider LLM inference (Ollama, OpenAI, OpenRouter, llama.cpp), multimodal ingestion (vision, audio transcription, speaker diarization), and knowledge graph. Rust + PostgreSQL.
NOASSERTION ★ 30 -
Give AI coding agents (Antigravity, Claude Code, Cursor) native macOS access to WhatsApp, iMessage, Call Recordings & Diarization, Calendar & Reminders, Apple Pay, Outlook, Wi-Fi diagnostics, Airbnb, browser automation, and reel downloading via single-file CLIs & MCP.
-
Self-hosted audio API in one Docker container. Stem separation, mastering, MIR (BPM/key/chords), audio→MIDI, text-to-music + SFX (MusicGen/Stable Audio/Riffusion/AudioLDM2), diarization, speech enhancement, restoration, embeddings + zero-shot tagging, workflows, async + webhooks. REST + MCP. CPU + CUDA.
oci WTFPL ★ 6 -
Private, offline speech-to-text for Linux: push-to-talk dictation into any app, meeting transcription with speaker diarization, and searchable history. Whisper, Parakeet and a local LLM on your GPU (Vulkan) or CPU. Native Omarchy and Hyprland integration, Wayland, MCP for AI agents.
GPL-3.0 ★ 2 -
音视频到转写文稿与结构化纪要的一条命令工具链:拆轨、人声分离、说话人分离、多引擎识别、字幕与 RTTM、规范 docx。默认离线,可选云端 ASR,支持续跑与批量,附 MCP 服务、跨会议检索与泳道图。 | One-command AV to transcript and minutes: demux, separation, diarization, multi-engine ASR, SRT/VTT/RTTM, docx. Offline-first, optional cloud ASR, resumable batches, MCP server and cross-meeting search.
MIT -
Open-source macOS meeting transcription with GPT, FluidAudio, Nemotron alignment and MCP
MIT -
Manifest Anything: wrap Plaud CLI and MCP for audio control, speaker identification, custom dictionaries, routing, and tasks.
MIT -
Modal podcast transcription pipeline (WhisperX + BGE + Chroma) with EU->US Chroma migration tooling
MIT -
mmmtrcactive
Local meeting transcription for humans and agents — a web app, an MCP server, and a CLI over one shared core.
npm MIT -
A full-stack multimodal meeting intelligence system with layered RAG, long-term memory, and skill-based generation.
MIT ★ 4 -
@dta.teks/wiumupdated this year
Local meeting transcription for humans and agents — a web app, an MCP server, and a CLI over one shared core.
npm MIT -
Open source meeting notes for macOS: no bot in the call, local Whisper transcription, who said what, summaries, ask across meetings, MCP server.
MIT -
A private, local-first macOS call recorder with Whisper transcription, speaker labeling, and a built-in MCP server for Codex.
MIT ★ 3 -
claude-meeting-mcpgithub.com/Aedelon/claude-meeting-mcpupdated this year
MCP Server to record, transcribe, and summarize any audio — meetings, YouTube, podcasts, lectures — with speaker identification and automatic meeting minutes. Cross-platform (macOS, Windows, Linux).
Apache-2.0 ★ 1 -
Model Context Protocol server that records calls on macOS (WhatsApp, Teams, Zoom, browser) and transcribes them locally with Whisper, with speaker separation.
MIT -
klaketgithub.com/huseyinstif/klaketupdated this year
Turn any video into LLM-ready data - transcripts, speaker labels, scenes, subtitles and on-screen text. Self-hostable, keyless, MCP-ready.
npm AGPL-3.0 ★ 3 -
Offline dictation and meeting notetaker for Windows: local Whisper transcription, speaker labels, and an MCP server for Claude Code.
MIT -
A private meeting archive for macOS: records calls, writes down who said what, keeps decisions and tasks. Local transcription, optional AI, built-in MCP server.
MIT ★ 6 -
whisper-valetgithub.com/yashkotha/whisper-valetupdated this year
The backstage crew for MacWhisper: drop an audio file, get restored audio + a speaker-labeled transcript back automatically. Demucs + pyannote + launchd automation with an MCP server for Claude.
MIT -
podcast-transcribergithub.com/a692570/podcast-transcriberupdated this year
Transcribe any podcast episode — CLI + MCP server. Three tiers: Gemini (speakers + tone), OpenAI Whisper API, or free local Whisper. Works on Mac, Windows, Linux.
MIT ★ 1 -
Backstory for macOS: record any call, get who said what and what was on screen, and give it to your agents over MCP. Local, no bot. Releases only.
-
OneMira C1 AI Voice Recorder — Official MCP Server & Mira CLI Integration for Cursor, Claude & Obsidian (Local AI · Speaker ID · Zero Cloud Subscription)
mcpb MIT -
Apex Vox — transcribe y resume reuniones con IA: oradores identificados, resúmenes, tareas y actas. FastAPI + Groq Whisper, PWA, extensión de Chrome y conector MCP para ChatGPT/Claude.
MIT -
Local meeting transcription MCP server: Whisper large-v3 + Qwen3-ASR on your own GPU, speaker labels, verify flags, optional Korean minutes. Audio never leaves your machine.
Apache-2.0 -
Voice dictation and meeting notes for Linux: local Parakeet speech-to-text, Claude cleanup, speaker-named transcripts with no bot in your calls, ask your meetings via CLI or MCP.
MIT ★ 1 -
Mac menu bar meeting recorder: records calls without bots, transcribes (cloud or on-device), and makes your meetings searchable from Claude via MCP.
MIT ★ 8 -
Sedjem: transcribe Egyptian Arabic–English meetings on your own computer, with speakers told apart by voice. Desktop app for Windows and Linux, plus benchmarks and notes.
AGPL-3.0 ★ 9 -
Speaker-labeled transcripts of recorded meetings, from the command line or an AI assistant (MCP). Transcription on-device with Parakeet or Nemotron 3.5, or via Meta Muse; speakers identified on-device with Nemotron 3. macOS on Apple Silicon.
MIT ★ 1 -
Records video calls (Zoom, Meet, Teams, any window), transcribes them locally with NVIDIA Nemotron, and gives AI clients the transcript and key frames over MCP. Fully local, no telemetry; it never summarizes.
MIT -
Local video editor: multitrack timeline with speed curves, masks, nesting and multicam, text-based editing with speaker separation, automatic cuts, captions and translated subtitles, vertical reframing, plain-language edit plans, NVENC renders and MCP tools.
MIT -
Record meetings in the browser, transcribe on free tiers, and let Claude write the minutes. Owners accept their own tasks, which sync to Zoho CRM and an Excel task plan. Cloudflare Workers + D1 + MCP.
MIT -
speaker-context-layergithub.com/sammyghe/speaker-context-layerupdated this year
Local, consented speaker attribution for AI assistants. An MCP server that answers who is speaking - with a voiceprint per language, and a match threshold you calibrate on your own voices. Voiceprints never leave your machine.
MIT ★ 1 -
Open-source multi-speaker video dubbing: transcribe, find who speaks when, translate, and re-voice every speaker with their own cloned voice, with timing fit, ducked background, dubbed MP4, multi-language audio track and SRT/VTT. CLI + MCP server + Agent Skill, voiced by Miragea.
pypi MIT -
Local speech-to-text for macOS (CLI + MCP) — whisper.cpp statically linked against Metal, so no API key, no ffmpeg, and no audio leaving the machine. Defaults to large-v3-turbo, measured on-machine to beat the Japanese-specialised alternative even on Japanese; VAD gating suppresses hallucinations over silence
MIT -
Local speech for Macs, Apple Silicon and Intel: text-to-speech with voice cloning and a studio voice, speech-to-text, system-wide dictation, and call recording with transcript, notes and speaker separation. Runs on-device with Metal. Daemon with an OpenAI-compatible HTTP API, a CLI and an MCP server.
Apache-2.0 ★ 1