Speech-to-Text MCP Server
Integrates with Hugging Face's pyannote.audio for speaker diarization, enabling identification and separation of different speakers in audio.
Integrates with OpenAI Whisper for automatic speech recognition, allowing transcription of audio files up to 60 minutes.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Speech-to-Text MCP Servertranscribe C:\meeting.mp3 with speaker diarization"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Speech-to-Text MCP Server
一个支持语音转文本和说话人分离的 Model Context Protocol (MCP) 服务器。
功能特性
✅ 支持上传最长 60 分钟的音频文件
✅ 自动语音识别 (使用 OpenAI Whisper)
✅ 说话人分离功能 (使用 pyannote.audio)
✅ 支持多种音频格式 (mp3, wav, m4a, flac 等)
✅ 输出带时间戳的转录文本
Related MCP server: whisper-telegram-mcp
系统要求
Python 3.10 或更高版本
Windows/Linux/macOS
至少 8GB RAM (推荐 16GB)
建议使用 GPU 以加速处理 (可选)
安装步骤
1. 创建虚拟环境
python -m venv venv
.\venv\Scripts\Activate.ps12. 安装依赖
pip install -e .3. 安装 FFmpeg (音频处理必需)
Windows:
下载 FFmpeg: https://ffmpeg.org/download.html
解压并添加到系统 PATH
或使用 Chocolatey:
choco install ffmpeg4. 获取 Hugging Face Token
pyannote.audio 需要 Hugging Face token:
创建一个新的 token
接受模型使用条款:
配置 MCP 客户端
在你的 MCP 客户端配置文件中添加此服务器:
{
"mcpServers": {
"speech-to-text": {
"command": "python",
"args": [
"e:\\demoProject\\speech_to_text\\server.py"
],
"env": {
"HUGGINGFACE_TOKEN": "your_token_here"
}
}
}
}使用方法
服务器提供以下工具:
1. transcribe_audio
转录音频文件为文本。
参数:
audio_file_path(必需): 音频文件的完整路径language(可选): 语言代码,如 "zh" (中文), "en" (英文), 默认自动检测enable_diarization(可选): 是否启用说话人分离,默认 false
示例调用:
{
"audio_file_path": "C:\\Users\\username\\Desktop\\meeting.mp3",
"language": "zh",
"enable_diarization": true
}2. get_supported_formats
获取支持的音频格式列表。
输出示例
不启用说话人分离:
[00:00:00.000 --> 00:00:05.000] 大家好,欢迎参加今天的会议。
[00:00:05.000 --> 00:00:12.000] 今天我们主要讨论项目进展。启用说话人分离:
[说话人 SPEAKER_00] [00:00:00.000 --> 00:00:05.000]
大家好,欢迎参加今天的会议。
[说话人 SPEAKER_01] [00:00:05.000 --> 00:00:12.000]
今天我们主要讨论项目进展。技术细节
语音识别引擎: OpenAI Whisper (medium 模型)
说话人分离: pyannote.audio 3.1
音频处理: pydub + FFmpeg
MCP 版本: 1.0
性能建议
对于长音频 (>30分钟),建议使用 GPU
首次运行会下载模型文件 (~1.5GB)
处理时间大约为音频时长的 0.5-2 倍 (取决于硬件)
故障排除
问题: FFmpeg 未找到
解决: 确保 FFmpeg 已安装并在 PATH 中
问题: CUDA out of memory
解决: 使用 CPU 模式或减少 batch size
问题: pyannote.audio 认证失败
解决: 检查 HUGGINGFACE_TOKEN 是否正确设置并接受了模型条款
许可证
MIT License
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseBqualityDmaintenanceA MCP server that enables transcription of audio files using OpenAI's Speech-to-Text API, with support for multiple languages and file saving options.Last updated1810MIT
- AlicenseAqualityDmaintenanceAn MCP server that enables transcribing local audio files and Telegram voice messages using OpenAI's Whisper via local inference or cloud API. It supports multiple audio formats, automatic language detection, and optional word-level timestamps for AI-powered audio analysis.Last updated51MIT
- FlicenseCqualityCmaintenanceAn MCP server that exposes speech-to-text and text-to-speech capabilities using a local speaches instance, allowing AI assistants to transcribe audio and generate speech.Last updated2
- AlicenseAqualityAmaintenanceMCP server for audio transcription using local faster-whisper or OpenAI Whisper API, enabling multilingual transcription with optional GPT post-processing.Last updated3MIT
Related MCP Connectors
MCP server exposing the AceDataCloud Fish Audio API (text-to-speech with voice conditioning)
MCP server for AI dialogue using various LLM models via AceDataCloud
An MCP server that integrates with Discord to provide AI-powered features.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/crackgo/speech-to-text-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server