Whissle MCP Server

Example Use Cases

• 1

  Transcribe audio files into text with word-level timestamps and optional boosted language model terms.

• 2

  Speaker diarization to produce per-speaker transcripts and segment timings.

• 3

  Translate text or generate concise summaries of long transcripts using an LLM-backed summarization tool.

Description

This MCP server provides programmatic access to Whissle's speech and text capabilities, including ASR (with timestamps and boosted language models), speaker diarization, text translation, and LLM-based summarization. It validates files, enforces size and format limits, and includes robust error handling and retry logic for common HTTP failures. The server requires a Whissle API authentication token and can be integrated into agent workflows or Claude via a configured MCP entry.