Time Series Filtering, Data Reduction and Voice Recognition in Communication Device

A computer implemented method for processing audio data, where one or more processors are programmed to perform steps comprising:

• obtaining time series audio data comprising audio data over a time period;
• partitioning the audio data in a plurality of time segments;
• transforming the audio data in the plurality of time segments into a plurality of feature values;
• identifying a subset of said plurality of feature values corresponding to a predetermined range of feature values;
• and
• storing said subset of said plurality of feature values corresponding to a predetermined or dynamically learned range of feature values to compress the time series audio data.

An apparatus for processing audio data to be transmitted over a data communication network, the apparatus comprising:

• a non-transitory computer readable medium configured to store computer executable programmed modules;
• a processor communicatively coupled with the non-transitory computer readable medium configured to execute programmed modules stored therein;
• an audio data module stored in the non-transitory computer readable medium and executable by the processor, said audio data module configured to receive time series audio data comprising audio data over a time period;
• a segment module stored in the non-transitory computer readable medium and executable by the processor, said segment module configured to partition the received time series audio data into a plurality of time segments;
• a transform module stored in the non-transitory computer readable medium and executable by the processor, said transform module configured to transform the audio data in the plurality of time segments into a plurality of feature values;
• a salience module stored in the non-transitory computer readable medium and executable by the processor, said salience module configured to identify a subset of said plurality of feature values corresponding to a predetermined or dynamically learned feature value range.