20 Best Speech & Transcription MCP Servers, Compared (November 2026)

The short answer

Which one, for your situation

Top MCP servers for Speech & Transcription

The ranking, with the evidence

• 1

  FunASR MCP Server

  Community favourite

  modelscope/FunASR ↗

  Best for: Local, private audio transcription: it runs FunASR and SenseVoice on your own machine and exposes a transcribe_audio tool rather than calling a hosted API.

  The repository documents the FunASR toolkit, a pip-installed Python library whose AutoModel class loads checkpoints such as Fun-ASR-Nano and SenseVoiceSmall and, with FSMN-VAD and CAM++ added, returns segments with timestamps and speaker indices, while its own description presents a transcribe_audio tool as the MCP entry point. Before installing, note that it assumes a local PyTorch and torchaudio environment (with a CUDA wheel matching the driver for GPU use), and that language and speaker coverage depends on the checkpoint chosen, with speaker indices local to a single recording rather than identified people.

  Score 71.8: show every number behind it

  • Adoption100 / 100 · weight 40%
    • GitHub stars100
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum71 / 100 · weight 14%
    • Stars gained, relative to size52
    • Stars gained, absolute100
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integrates100

  Weighted mean of the five

  92.1

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  71.8

• 2

  NeuroLink

  Community favourite

  juspay/neurolink ↗

  Best for: TypeScript apps that need speech-to-text or text-to-speech alongside multi-provider LLM calls: it documents voice support and includes a built-in MCP client for connecting tool servers.

  NeuroLink is an npm-published TypeScript SDK and CLI that provides one interface to multiple LLM providers and a built-in MCP client for stdio, HTTP, SSE, and WebSocket connections, with documented streaming, tool calling, RAG, voice TTS/STT/realtime, and task scheduling. Before choosing it, note that it ships as an npm package and TypeScript SDK/CLI, so adoption assumes an npm and TypeScript toolchain.

  Score 61.3: show every number behind it

  • Adoption62 / 100 · weight 40%
    • GitHub stars56
    • npm downloads40
      downloads show none of the weekday rhythm human traffic has; halved
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum84 / 100 · weight 14%
    • Stars gained, relative to size96
    • Stars gained, absolute51
    • npm download trend100
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integrates100

  Weighted mean of the five

  78.6

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  61.3

• 3

  leemysw/yovoice

  Community favourite

  leemysw/yovoice ↗

  Best for: Local speech generation, voice cloning, and score rendering without cloud API calls: it exposes upload_voice, submit_generation, get_generation, and render_score tools.

  The server exposes nine MCP tools for local speech generation, voice cloning, voice import, model and voice listing, task status and cancellation, and score rendering to WAV. Before generation, a model must be installed locally and selected through list_models.

  Score 59.3: show every number behind it

  • Adoption83 / 100 · weight 40%
    • GitHub stars83
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum≈0 / 100 · weight 14%
    • Stars gained, relative to sizenot measured
      no snapshot history yet
    • Stars gained, absolutenot measured
      no snapshot history yet
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality76 / 100 · weight 13%
    • Tool description quality68
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  76.0

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  59.3

• 4

  claude-real-video

  Community favourite

  HUANGCHIHHUNGLeo/claude-real-video ↗

  Best for: For letting an LLM inspect a video URL or local file: its watch_video tool returns scene-aware keyframes with a timestamped transcript, processed locally.

  The server is documented as providing a watch_video tool that takes a URL or local file and returns scene-aware keyframes with a timestamped transcript, alongside a crv CLI and local crv-web interface for producing frames and transcripts. It assumes Python 3.10+ and ffmpeg on the local machine for video processing.

  Score 53.5: show every number behind it

  • Adoption89 / 100 · weight 40%
    • GitHub stars84
    • PyPI downloads31
      PyPI downloads show no weekday rhythm; halved
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum83 / 100 · weight 14%
    • Stars gained, relative to size79
    • Stars gained, absolute79
    • PyPI download trend97
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  89.2

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  53.5

• 5

  Whispera

  Steady

  sapoepsilon/Whispera ↗

  Best for: Dictating into any text field on an Apple Silicon Mac: a global shortcut records speech, transcribes it on-device with WhisperKit, and pastes at the cursor.

  Whispera is a native macOS menu bar app rather than an installable package: its README documents a dictation shortcut that inserts transcripts into the focused text field, a queue for transcribing audio files, media URLs and YouTube videos, searchable history, and optional LLM post-processing. Choosing it means running macOS 14 or later on Apple Silicon, since Intel Macs are unsupported, and installing from a DMG or Homebrew cask because no npm or PyPI package is published.

  Score 52.9: show every number behind it

  • Adoption56 / 100 · weight 40%
    • GitHub stars56
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance95 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade80
    • Commit cadence100
  • Momentum33 / 100 · weight 14%
    • Stars gained, relative to size39
    • Stars gained, absolute24
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  67.8

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  52.9

• 6

  Vexa

  Community favourite

  Vexa-ai/vexa ↗

  Best for: AI tools needing meeting transcripts over MCP: Vexa sends bots into Google Meet, Teams, or Zoom calls and exposes transcript retrieval through its API.

  Vexa documents an API for sending a bot into Google Meet, Microsoft Teams, or Zoom calls and reading speaker-attributed transcripts, plus an agent chat endpoint for querying captured meetings. The hosted service covers bots and transcription, while the agent chat and sandboxed knowledge plane require self-hosting the full stack.

  Score 51.7: show every number behind it

  • Adoption87 / 100 · weight 40%
    • GitHub stars87
    • npm downloadsnot measured
      npm publishes vexa from pan-alexey/vexa
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance95 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade80
    • Commit cadence100
  • Momentum76 / 100 · weight 14%
    • Stars gained, relative to size73
    • Stars gained, absolute80
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integrates100

  Weighted mean of the five

  86.2

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  51.7

• 7

  Voice Mode

  Community favourite

  mbailey/voicemode ↗

  Best for: When typing is impractical, such as walking or cooking: it documents voice conversation setup with Claude Code, including local Whisper speech-to-text and Kokoro text-to-speech.

  Voice Mode documents installation of an MCP server for voice conversations with Claude Code, via a Claude Code plugin or a PyPI launcher, with optional local Whisper speech-to-text and Kokoro text-to-speech. It assumes a computer with a microphone and speakers, and running voice services locally requires separate Whisper.cpp and Kokoro setup.

  Score 51.5: show every number behind it

  • Adoption84 / 100 · weight 40%
    • GitHub stars79
    • PyPI downloads33
      PyPI downloads show no weekday rhythm; halved
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance95 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade80
    • Commit cadence100
  • Momentum81 / 100 · weight 14%
    • Stars gained, relative to size77
    • Stars gained, absolute70
    • PyPI download trend100
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  85.8

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  51.5

• 8

  screenpipe

  Community favourite

  screenpipe/screenpipe ↗

  Best for: Meeting and workstation recall for AI agents: it continuously records screen, audio, transcription, speakers, OCR, and app switches locally and supports natural-language search.

  Its MCP tool list has not been inspected; the README documents a desktop app and CLI that continuously record screen, audio, accessibility-tree text, OCR fallback, transcription, speakers, keyboard input, and app switches into local storage, plus setup for Claude. It requires completing setup and leaving a local recorder running before an agent can query history; local capture and search work offline after setup, while cloud AI, sync, and connected services require network access.

  Score 50.9: show every number behind it

  • Adoption100 / 100 · weight 40%
    • GitHub stars100
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum74 / 100 · weight 14%
    • Stars gained, relative to size57
    • Stars gained, absolute100
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust15 / 100 · weight 9%
    • License0
    • Published by the vendor it integrates100

  Weighted mean of the five

  84.8

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  50.9

• 9

  omniseek

  Steady

  Battam1111/omniseek ↗

  Best for: Transcribing spoken content from video, podcast, or audio URLs: it exposes a local SenseVoice ASR tool that is keyless and caches transcripts.

  OmniSeek exposes 18 MCP tools, including omniseek_search for curated sources, omniseek_read for URLs and document files, omniseek_transcribe for speech through local SenseVoice ASR, omniseek_view for images and video frames, and tools for papers, authors, graphs, and monitoring. Before relying on it for speech transcription, note that podcast and video transcription is listed as an optional add-on installed separately.

  Score 50.7: show every number behind it

  • Adoption55 / 100 · weight 40%
    • GitHub stars50
    • PyPI downloads32
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum62 / 100 · weight 14%
    • Stars gained, relative to size76
    • Stars gained, absolute39
    • PyPI download trendnot measured
      no download history for the selected registry
  • Tool quality100 / 100 · weight 13%
    • Tool description quality93
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  76.5

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  50.7

• 10

  SuperCMO Skills

  Community favourite

  SupercmoHQ/superCMO-skills ↗

  Best for: Captioning existing marketing videos: transcribe returns word-level timestamps, and caption_video burns styled social captions from a word-timed transcript.

  SuperCMO Skills exposes 22 tools for image, video and audio generation, transcription with word timestamps, captioning and overlay of video, social research, and calendar scheduling. It requires media-generation keys, either the user's own or paid access through SuperCMO, and its setup_status tool is meant to be called first to see which image, video and audio capabilities are ready.

  Score 50.5: show every number behind it

  • Adoption68 / 100 · weight 40%
    • GitHub stars62
    • PyPI downloads45
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance95 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade80
    • Commit cadence100
  • Momentum99 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute98
    • PyPI download trend100
  • Tool quality85 / 100 · weight 13%
    • Tool description quality77
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integrates100

  Weighted mean of the five

  84.2

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  50.5

• 11

  ElevenLabs MCP Server

  Steady

  mamertofabian/elevenlabs-mcp-server ↗

  Best for: Multi-voice narration for scripts and podcasts: it generates audio from structured scripts with multiple voices and actors, plus single-voice text-to-speech.

  The server wraps the ElevenLabs text-to-speech API in six tools: generate_audio_simple and generate_audio_script for synthesis, list_voices for voice metadata, plus get_audio_file, get_voiceover_history and delete_job for managing generated audio jobs. It requires an ELEVENLABS_API_KEY environment variable and runs via uvx with no separate install, so credentials must be configured before the first call.

  Score 50.3: show every number behind it

  • Adoption52 / 100 · weight 40%
    • GitHub stars52
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance83 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade80
    • Commit cadence40
  • Momentum0 / 100 · weight 14%
    • Stars gained, relative to size0
    • Stars gained, absolute0
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality73 / 100 · weight 13%
    • Tool description quality65
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  59.2

  × relevance: the keyword is dedicated here

  1.00

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  50.3

• 12

  VideoNote-MCP

  Emerging

  HuangYincan/VideoNote-MCP ↗

  Best for: Turning Bilibili, YouTube, Douyin, Kuaishou, TikTok, or local video/audio into Markdown notes: it exposes generate_note and prepare_note_material with transcription and task tools.

  VideoNote-MCP exposes tools that accept Bilibili, YouTube, Douyin, Kuaishou, TikTok, or local media paths, transcribe them through configurable engines, manage asynchronous tasks, and produce Markdown notes or material packs for an agent to write notes from. Before choosing it, note that setup requires uv and FFmpeg, and LaTeX or Typst PDF output depends on compilers, fonts, and dependencies prepared locally rather than on the MCP server.

  Score 49.8: show every number behind it

  • Adoption50 / 100 · weight 40%
    • GitHub stars50
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum82 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute56
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality81 / 100 · weight 13%
    • Tool description quality73
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  75.1

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  49.8

• 13

  Jarvis Assistant Vocal MCP Server

  Community favourite

  sosoj92/jarvis-assistant-vocal ↗

  Best for: French-speaking Windows users running the local Jarvis voice assistant: its MCP server exposes Jarvis's home-automation and PC-control tools to MCP clients.

  Jarvis Assistant Vocal MCP Server exposes the home-automation and PC-control tools of the local Jarvis voice assistant to MCP clients such as Claude Desktop or Hermes. It targets Windows 11 and assumes a local Jarvis setup with a microphone, plus a provider API key in cloud mode.

  Score 49.6: show every number behind it

  • Adoption68 / 100 · weight 40%
    • GitHub stars68
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum94 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute84
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  82.7

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  49.6

• 14

  AudacityMCP

  Emerging

  xDarkzx/Audacity-MCP ↗

  Best for: Cleaning up spoken-word recordings and podcasts locally in Audacity: it exposes 144 tools, including noise reduction, silence trimming, compression, and one-click podcast cleanup.

  AudacityMCP connects MCP-compatible AI assistants to Audacity 3.x through a named-pipe interface, exposing 144 tools for selection-based editing, effects, analysis, cleanup, mastering, and automated pipelines. It cannot control Audacity 4, which removed mod-script-pipe, and local audio processing requires Audacity 3.x with that module enabled.

  Score 49.2: show every number behind it

  • Adoption53 / 100 · weight 40%
    • GitHub stars53
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance97 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence85
  • Momentum84 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute61
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality68 / 100 · weight 13%
    • Tool description quality60
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  74.2

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  49.2

• 15

  guaardvark

  Community favourite

  guaardvark/guaardvark ↗

  Best for: Local media generation from an MCP client rather than transcription: it exposes image, video, animation and music-video tools that queue renders and return batch IDs.

  Guaardvark exposes 55 tools for local media generation, including image, video, animation and music-video workflows, plus document and code search, memory and Linux media-player controls. Its listed tools do not include an audio transcription tool, and video generation assumes a Linux machine with an NVIDIA GPU.

  Score 49.0: show every number behind it

  • Adoption64 / 100 · weight 40%
    • GitHub stars62
    • npm downloadsnot measured
      no npm package
    • Used through Glama14
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum88 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute69
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality83 / 100 · weight 13%
    • Tool description quality75
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  81.7

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  49.0

• 16

  mcp-video-analyzer

  Emerging

  guimatheus92/mcp-video-analyzer ↗

  Best for: For transcript-plus-frame analysis of Loom or direct video URLs: it exposes tools for transcripts, key frames, OCR, metadata, and annotated timelines.

  The server exposes eight tools for extracting transcripts, key frames, OCR text, metadata, comments, and annotated timelines from video sources. Before choosing it, note that major platform URLs such as YouTube, Vimeo, TikTok, Instagram, X, Twitch, Dailymotion, and Facebook require yt-dlp, while direct URLs and local files can still get frames without it.

  Score 47.9: show every number behind it

  • Adoption56 / 100 · weight 40%
    • GitHub stars51
    • npm downloads30
      downloads show none of the weekday rhythm human traffic has; halved
    • Used through Glama33
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum87 / 100 · weight 14%
    • Stars gained, relative to size100
    • Stars gained, absolute57
    • npm download trend100
  • Tool quality93 / 100 · weight 13%
    • Tool description quality85
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  79.8

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  47.9

• 17

  kesha-voice-kit

  Steady

  drakulavich/kesha-voice-kit ↗

  Best for: Local, offline voice workflows through an MCP client: it exposes transcribe, synthesize, and list tools and runs models on-device without API keys.

  Kesha Voice Kit runs an MCP server, kesha mcp, that exposes transcribe, synthesize, and list tools to MCP clients such as Claude, Cursor, Codex, and Gemini. Before it can transcribe, it requires Bun 1.4.0 or newer and an explicit kesha install model download of about 2.5 GB on Linux/Windows or about 0.6 GB on Apple Silicon.

  Score 46.2: show every number behind it

  • Adoption51 / 100 · weight 40%
    • GitHub stars48
    • npm downloads22
      downloads show none of the weekday rhythm human traffic has; halved
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum52 / 100 · weight 14%
    • Stars gained, relative to size45
    • Stars gained, absolute23
    • npm download trend100
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  69.7

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  46.2

• 18

  Fast MCP Telegram

  Steady

  leshchenko1979/fast-mcp-telegram ↗

  Best for: Handling voice notes that arrive in Telegram chats: the server transcribes voice messages and exposes eight tools for searching, reading and replying across chats.

  Fast MCP Telegram sits between an MCP client and Telegram's MTProto API, exposing eight tools: global and per-chat message search, chat discovery and profile lookup, sending and editing messages, sending to a phone number, and a low-level invoke_mtproto call for methods the wrappers do not cover. Choosing it means giving the server a Telegram login, by QR scan or phone with code, and keeping that session available, since every tool acts as the authenticated account and edit_message only works on messages that account sent.

  Score 45.8: show every number behind it

  • Adoption46 / 100 · weight 40%
    • GitHub stars43
    • PyPI downloads21
      PyPI downloads show no weekday rhythm; halved
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance100 / 100 · weight 24%
    • Last commit touching this server100
    • Repository maintenance grade100
    • Commit cadence100
  • Momentum45 / 100 · weight 14%
    • Stars gained, relative to size35
    • Stars gained, absolute16
    • PyPI download trend100
  • Tool quality88 / 100 · weight 13%
    • Tool description quality80
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  69.1

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: modest but real audience

  0.85

  Composite score

  45.8

• 19

  Ableton MCP Extended

  Abandoned but popular

  uisato/ableton-mcp-extended ↗

  Best for: Controlling an Ableton Live session by natural language: it documents session, track, clip, device, browser, and ElevenLabs audio actions through an installed remote script.

  The server documents an MCP interface that translates natural language into Ableton Live actions through an installed remote script, covering session and transport control, track and MIDI clip editing, device parameters, browser loading, and ElevenLabs text-to-speech import. Before choosing it, note that it requires Ableton Live 11+ and Python 3.10+, along with manual Remote Script installation and Control Surface configuration.

  Score 45.2: show every number behind it

  • Adoption61 / 100 · weight 40%
    • GitHub stars61
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance40 / 100 · weight 24%
    • Last commit touching this server68
    • Repository maintenance grade5
    • Commit cadence5
  • Momentum42 / 100 · weight 14%
    • Stars gained, relative to size47
    • Stars gained, absolute33
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality≈70 / 100 · weight 13%
    • Tool description quality≈70
      tool descriptions not yet scored
    • Built and inspected by Glamanot measured
      never built and inspected by Glama
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integratesnot measured
      not published by the vendor it integrates

  Weighted mean of the five

  57.9

  × relevance: the keyword is declared here

  0.78

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  45.2

• 20

  MiniMax MCP Server

  Community favourite

  MiniMax-AI/MiniMax-MCP ↗

  Best for: When an MCP client needs text-to-speech or voice cloning: it exposes text_to_audio and voice_clone tools.

  MiniMax MCP Server exposes six tools for text-to-audio conversion, voice listing, voice cloning, audio playback, video generation, and image generation. It requires a MiniMax API key and an installed uv before setup.

  Score 44.3: show every number behind it

  • Adoption80 / 100 · weight 40%
    • GitHub stars80
    • npm downloadsnot measured
      no npm package
    • Used through Glamanot measured
      not used through Glama in the last 30 days
  • Maintenance69 / 100 · weight 24%
    • Last commit touching this server88
    • Repository maintenance grade30
    • Commit cadence65
  • Momentum42 / 100 · weight 14%
    • Stars gained, relative to size37
    • Stars gained, absolute48
    • npm download trendnot measured
      no download history for the selected registry
  • Tool quality81 / 100 · weight 13%
    • Tool description quality73
    • Built and inspected by Glama100
  • Trust100 / 100 · weight 9%
    • License100
    • Published by the vendor it integrates100

  Weighted mean of the five

  73.8

  × relevance: the keyword is tagged here

  0.60

  × continuity: actively changing

  1.00

  × evidence: widely adopted

  1.00

  Composite score

  44.3

Questions people ask

Other comparisons

• Best Figma MCP servers
• Best Obsidian MCP servers
• Best YouTube MCP servers
• Best Jira MCP servers
• Best PostgreSQL MCP servers
• Best MySQL MCP servers
• Best GitHub MCP servers
• Best Notion MCP servers
• Best Linear MCP servers
• Best Discord MCP servers
• Best MongoDB MCP servers
• Best CRM MCP servers
• Best Payments MCP servers
• Best Ecommerce MCP servers
• Best Social Media MCP servers
• Best Browser Automation MCP servers
• Best Email MCP servers
• Best PDF MCP servers
• Best Kubernetes MCP servers
• Best Docker MCP servers
• Best AWS MCP servers
• Best Spreadsheets MCP servers
• Best SEO MCP servers
• Best Web Search MCP servers
• Best Web Scraping MCP servers
• Best Vector Database MCP servers
• Best Code Execution MCP servers
• Best Calendar MCP servers
• Best Observability MCP servers
• Best Image Generation MCP servers
• Best Maps MCP servers
• Best Travel MCP servers
• Best Stock Market MCP servers
• Best Web Analytics MCP servers
• Best Memory MCP servers
• Best Documentation MCP servers
• Best Security MCP servers