MACPI: Mining Audio Cues from Voice Data

MITRE’s innovative Mining Audio Cues from Voice Data (MACPI) tool provides a new method of analysis of voice data and emotional speech classification. Its analytic capability can improve the efficiency, consistency, and accuracy of diagnosis of some health conditions by analysis of data from voice recordings. MACPI enables automatic keyword indexing, phonetic indexing, and audio feature extraction. The search and retrieval functions provide quick access to specific segments of voice recordings for consistent analysis and verification.

The tool’s audio processing capability also enables detection of non-verbal cues such as speech rate and elevated speech captured in recordings. MACPI was originally developed as a method to aid clinicians in the diagnosis of post-traumatic stress disorder. 

For more detailed information on MITRE’s MACPI or licensing options, contact the MITRE Technology Transfer Office at techtransfer@mitre.org.