Audio Analyzer

What it does

The Audio Analyzer skill provides a toolkit for analyzing audio files and extracting detailed information about their characteristics. It enables AI agents to determine tempo (BPM), musical key, frequency content, loudness metrics (RMS, peak, LUFS), and generate visualizations of the audio data. This allows for in-depth understanding and reporting on various aspects of an audio track.

When to use it

• Analyzing music tracks to identify genre or mood based on tempo and key.
• Evaluating audio recordings for loudness compliance (e.g., LUFS) for broadcast or streaming platforms.
• Generating visual representations (waveforms, spectrograms) of audio for educational or presentation purposes.
• Identifying dominant frequencies in an audio sample to understand its sonic characteristics.

Key capabilities

• Tempo/BPM Detection: Accurately determines the beats per minute with a confidence score.
• Key Detection: Identifies the musical key and mode (major/minor) of the audio.
• Frequency Analysis: Provides spectrum, dominant frequencies, and frequency band information.
• Loudness Metrics: Calculates RMS, peak levels, LUFS, and dynamic range.
• Visualization: Generates waveform plots, spectrograms, chromagrams, and beat grids.
• Report Generation: Exports analysis results in JSON format.

Example prompts

• "Analyze this audio file ('song.mp3') and tell me the BPM."
• "Can you identify the musical key of this recording?"
• "Generate a spectrogram visualization for 'audio.wav'."
• "What is the LUFS value of this track, and is it broadcast-ready?"

Tips & gotchas

• The skill requires Python to be installed and accessible within the AI agent's environment.
• You can specify a custom sample rate when initializing the AudioAnalyzer if needed (e.g., analyzer = AudioAnalyzer("audio.wav", sr=44100)).
• Individual analysis methods (tempo, key, loudness) can be run separately for more targeted results.