mcp-audio-analysis

Instructions

Computes the MFCC of the given audio time series using librosa.
The MFCC is a representation of the audio signal in terms of its
spectral content, which is useful for music analysis.
The MFCC is computed using the following parameters:
- path_audio_time_series_y: The path to the audio time series (CSV file).
    It's sometimes better to take harmonics only

Implementation Reference

• src/mcp_music_analysis/server.py:146-165 (handler)

  The handler function for the 'mfcc' tool. It loads the audio time series from a CSV file, computes the MFCC features using librosa.feature.mfcc, saves the result to a new CSV file in temp dir, and returns the path to that file. The @mcp.tool() decorator registers it as an MCP tool.

  @mcp.tool()
  def mfcc(
      path_audio_time_series_y: str,
  ) -> str:
      """
      Computes the MFCC of the given audio time series using librosa.
      The MFCC is a representation of the audio signal in terms of its
      spectral content, which is useful for music analysis.
      The MFCC is computed using the following parameters:
      - path_audio_time_series_y: The path to the audio time series (CSV file).
          It's sometimes better to take harmonics only
      """
      y = np.loadtxt(path_audio_time_series_y, delimiter=";")
      mfcc = librosa.feature.mfcc(y=y)
      # Save the mfcc to a CSV file
      name = path_audio_time_series_y.split("/")[-1].split(".")[0] + "_mfcc"
      mfcc_path = os.path.join(tempfile.gettempdir(), name + ".csv")
      np.savetxt(mfcc_path, mfcc, delimiter=";")
      # Return the path to the CSV file
      return mfcc_path