Adaptive vehicle state-based hands-free phone noise reduction with learning capability

A system comprising:

• (i) a vehicle comprising:
• (a) a microphone configured to capture audio;
• (b) a memory configured to store a Weiner filter and a database including a plurality of predetermined discrete prefilters and a plurality of selection strategies, each selection strategy referencing at least one prefilter;
• (c) processor(s) in communication with the memory, the processor(s) configured to, while a hands free phone system is operating:
• receive vehicle operational state(s), external information, and feedback data;
• select at least one of the selection strategies from the database based on the vehicle operational state(s), the external information, and the feedback data;
• apply the prefilter(s) referenced by the selected selection strategy to the captured audio;
• apply the Weiner filter to the prefiltered audio;
• and
• update the selected selection strategy based on an analysis;
• and
• (ii) server(s) configured to:
• generate the analysis by analyzing a quality of the Weiner filtered audio to change the stored selection strategy while the hands free phone system is operating;
• segment the Weiner filtered audio into first segments determined to include human speech and into second segments determined to lack human speech;
• and
• analyze the quality of the Weiner filtered audio by comparing at least one of the first segments to at least one of the second segments.