transcript-mcp-claude
An all-in-one MCP server for transcription, supporting YouTube, audio, and video with translation, summarization, chapter generation, subtitle export, and batch processing via 19 tools.
README
transcript-mcp-claude
MCP Server All-in-One สำหรับ Transcription - รวม YouTube, Audio, Video + Translation, Summarization, Chapters, Subtitles, Batch Processing
Features
19 MCP Tools ใน 8 หมวด:
| หมวด | Tools | คำอธิบาย |
|---|---|---|
| YouTube (3) | youtube_to_text, youtube_video_info, youtube_list_captions |
ดึง transcript และข้อมูลจาก YouTube |
| Audio (2) | audio_to_text, audio_info |
แปลงไฟล์เสียงเป็นข้อความด้วย Whisper |
| Video (2) | vdo_to_text, vdo_info |
แปลงไฟล์วิดีโอเป็นข้อความด้วย ffmpeg + Whisper |
| Translation (2) | translate_transcript, create_multilingual_subtitles |
เตรียมข้อมูลสำหรับแปลภาษา |
| Summarization (3) | summarize_transcript, extract_key_moments, generate_blog_post |
สรุปเนื้อหา, ดึง key moments, สร้าง blog post |
| Chapters (2) | generate_chapters, format_chapters_youtube |
สร้าง chapter markers + YouTube format |
| Subtitles (3) | export_subtitles, convert_subtitle_format, adjust_subtitle_timing |
Export SRT/VTT/ASS, แปลง format, ปรับ timing |
| Batch (2) | batch_transcribe, transcribe_folder |
ประมวลผลหลายไฟล์พร้อมกัน |
Tech Stack
- Language: TypeScript (strict mode, ES2022)
- Runtime: Node.js 22 + Python 3.12
- MCP SDK: @modelcontextprotocol/sdk v1.0.0
- Transport: Streamable HTTP (port 3013) + Stdio
- STT Engine: faster-whisper (CTranslate2, CPU, int8)
- YouTube: Innertube API + yt-dlp fallback
- Video/Audio: ffmpeg + ffprobe
- Build: Multi-stage Docker (node:22-slim → python:3.12-slim)
Quick Start
Docker (แนะนำ)
# Build and start
docker compose up -d --build
# ตรวจสอบ
curl http://localhost:3013/health
Local Development
# Install dependencies
npm install
# Build TypeScript
npm run build
# Start HTTP server
npm start
# Start Stdio server
npm run start:stdio
# Development mode (watch + auto-compile)
npm run dev
Environment Variables
| Variable | Default | คำอธิบาย |
|---|---|---|
PORT |
3013 | Server port |
HOST |
0.0.0.0 | Server host |
DEFAULT_LANG |
th | ภาษาเริ่มต้น |
WHISPER_MODEL |
tiny | Whisper model: tiny, base, small |
WHISPER_COMPUTE_TYPE |
int8 | Compute type: int8, float16, float32 |
MAX_AUDIO_DURATION |
600 | Audio max duration (วินาที) |
MAX_VDO_DURATION |
1800 | Video max duration (วินาที) |
OMP_NUM_THREADS |
2 | CPU threads สำหรับ inference |
MCP Configuration
เพิ่มใน .mcp.json:
{
"mcpServers": {
"transcript": {
"url": "http://localhost:3013/mcp"
}
}
}
Usage Examples
YouTube Transcript
ดึง transcript จาก YouTube video นี้: https://youtube.com/watch?v=xxx
Audio Transcription
แปลงไฟล์เสียง /app/data/audio.mp3 เป็นข้อความ
Video Transcription
แปลงไฟล์วิดีโอ /app/data/video.mp4 เป็นข้อความ
Export Subtitles
สร้างไฟล์ SRT จาก transcript ที่ได้
Generate Chapters
สร้าง chapter markers จาก transcript สำหรับใส่ YouTube description
Batch Processing
Transcribe ไฟล์ทั้งหมดใน folder /app/data/
Supported Formats
- Audio: MP3, WAV, M4A, OGG, FLAC, WebM, WMA, AAC, Opus
- Video: MP4, MKV, AVI, MOV, WebM, FLV, WMV, TS, M4V
- Subtitles: SRT, VTT, ASS, TXT, JSON
Architecture
Client (Claude) → MCP Protocol → transcript-mcp-claude (port 3013)
├── YouTube (Innertube API / yt-dlp)
├── Audio (faster-whisper)
├── Video (ffmpeg → faster-whisper)
├── Translation (data structuring)
├── Summarization (data structuring)
├── Chapters (timestamp analysis)
├── Subtitles (SRT/VTT/ASS export)
└── Batch (multi-file orchestration)
Port Convention
| Port | Project |
|---|---|
| 3010 | youtube-mcp-claude |
| 3011 | audio-mcp-claude |
| 3012 | vdo-mcp-claude |
| 3013 | transcript-mcp-claude (All-in-One) |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.