Audioscrape

Adicionar pesquisa de áudio via MCP - Pesquise qualquer áudio em segundos

Ver em audioscrape.com

Documentação

Add audio search to Claude - Search any audio in seconds

Server Active

What is this?

This is the official Audioscrape Remote MCP (Model Context Protocol) server endpoint. It enables AI assistants like Claude to directly access and search over 170,000 hours of transcribed audio, including podcasts, interviews, and public meetings, through a standardized protocol.

Features

  • Search over 170,000 hours of transcribed audio
  • Access detailed transcriptions with timestamps
  • Find specific moments in podcasts and interviews
  • Browse content by topic, speaker, or keyword
  • Real-time semantic search across spoken content

How to Connect

To integrate this MCP server with Claude Desktop or other compatible AI tools, you'll need to configure your MCP client with our server endpoint.

Technical Details

Protocol Version: MCP 2025-11-25 (latest)

Server Endpoint: https://mcp.audioscrape.com

Authentication: OAuth 2.1 with PKCE (mandatory S256)

Transport: HTTP with JSON-RPC 2.0

OAuth Discovery Endpoints:

Authorization Server: /.well-known/oauth-authorization-server

Protected Resource: /.well-known/oauth-protected-resource

Client Registration: POST /oauth/register (RFC 7591)

Resource Indicators: Supported (RFC 8707)