Best MCP Servers for Transcription & Voice (2026)
Speech in, speech out: YouTube transcript extraction (two different servers), Voice MCP and Kokoro for text-to-speech, MarkItDown for turning audio and documents into Markdown, and Content Core for extracting transcripts and text from media files.
This page collects the 6 MCP servers in our directory that genuinely cover transcription & voice — together they have 123,187 GitHub stars, and 0 of them run as hosted services you can connect to without installing anything. The highest-rated options are microsoft/markitdown, mbailey/voice-mcp, kimtaeyoon83/mcp-server-youtube-transcript.
How to choose
Use a YouTube transcript server when the source is a video link, Kokoro or Voice MCP for generated speech, and MarkItDown when transcription is one step in a document pipeline.
microsoft/markitdown
by microsoft
MarkItDown is Microsoft's lightweight Python utility for converting files to Markdown for use with LLMs and text-analysis pipelines, and this repository includes the markitdown-mcp package that exposes the converter through MCP. It converts PDF, PowerPoint, Word, Excel, images (EXIF metadata and OCR), audio (EXIF metadata and speech transcription), HTML, text-based formats such as CSV, JSON, and XML, ZIP files, YouTube URLs, and EPUBs, preserving document structure such as headings, lists, tables, and links.
npx -y microsoft-markitdownmbailey/voice-mcp
by mbailey
🐍 🏠 - Complete voice interaction server supporting speech-to-text, text-to-speech, and real-time voice conversations through local microphone, OpenAI-compatible APIs, and LiveKit integration
uvx voice-mode-installkimtaeyoon83/mcp-server-youtube-transcript
by kimtaeyoon83
📇 ☁️ - Fetch YouTube subtitles and transcripts for AI analysis
npx -y -y @smithery/clianaisbetts/mcp-youtube
by anaisbetts
📇 ☁️ - Fetch YouTube subtitles
npx -y @anaisbetts/mcp-youtubelfnovo/content-core
by lfnovo
🐍 🏠 - Extract content from URLs, documents, videos, and audio files using intelligent auto-engine selection. Supports web pages, PDFs, Word docs, YouTube transcripts, and more with structured JSON r
uvx content-coremberg/kokoro-tts-mcp
by mberg
MCP server that converts text to speech using the open-weight Kokoro TTS model (kokoro-onnx weights). It generates MP3 files (converting WAV output with ffmpeg) saved to a local folder, with an option to auto-upload them to an S3 bucket, and is configured through environment variables for voice, speed, language, and storage.
npx -y mberg-kokoro-tts-mcpFrequently asked questions
What is the most popular MCP server for Transcription & Voice?
microsoft/markitdown leads this list with 120,761 GitHub stars. MarkItDown is Microsoft's lightweight Python utility for converting files to Markdown for use with LLMs and text-analysis pipelines, and this repository includes the markitdown-mcp package that…
Are these Transcription & Voice MCP servers free to use?
6 of the 6 servers here publish a license in the directory (MIT, Apache-2.0). 0 of them run as hosted services you can connect to remotely; the rest you install and run yourself. Hosted services may have their own pricing.
How do I choose an MCP server for Transcription & Voice?
Use a YouTube transcript server when the source is a video link, Kokoro or Voice MCP for generated speech, and MarkItDown when transcription is one step in a document pipeline.