brainiall-mcp-server

Instructions

Implementation Reference

• server.py:94-115 (handler)

  Main handler function for the synthesize_speech tool. It converts text to speech by making an async HTTP POST request to the Brainiall API endpoint '/v1/tts/synthesize' with text, voice (default: 'alloy'), and speed (default: 1.0) parameters. Returns base64-encoded MP3 audio.

  @mcp.tool()
  async def synthesize_speech(
      text: Annotated[str, "The text to convert to speech"],
      voice: Annotated[str, "Voice ID to use (see list_voices for available options)"] = "alloy",
      speed: Annotated[float, "Speaking speed multiplier (0.5 to 2.0)"] = 1.0,
  ) -> dict:
      """Convert text to natural-sounding speech audio.
  
      Returns base64-encoded audio in MP3 format.
      Use list_voices to see available voice options.
      """
      async with _client() as client:
          response = await client.post(
              "/v1/tts/synthesize",
              json={
                  "text": text,
                  "voice": voice,
                  "speed": speed,
              },
          )
          response.raise_for_status()
          return response.json()

• server.py:94-99 (schema)

  Input schema definition for synthesize_speech using Annotated type hints. Parameters: text (str, required), voice (str, default 'alloy'), and speed (float, range 0.5-2.0, default 1.0). Each parameter includes descriptive metadata for tool documentation.

  @mcp.tool()
  async def synthesize_speech(
      text: Annotated[str, "The text to convert to speech"],
      voice: Annotated[str, "Voice ID to use (see list_voices for available options)"] = "alloy",
      speed: Annotated[float, "Speaking speed multiplier (0.5 to 2.0)"] = 1.0,
  ) -> dict:

• server.py:94-94 (registration)

  Tool registration decorator @mcp.tool() that registers the synthesize_speech function as an MCP tool with the FastMCP server instance.

  @mcp.tool()