Data recognition and separation engine

Abstract

Embodiments disclosed herein extend to methods, systems, and computer program products for analyzing digital data. A source of digital data is analyzed and separated into segments, each segment having an identifiable characteristic. The separated segments are copied into planes of a higher dimension. The separated segments are compared to determine a resemblance factor. A fingerprint is generated for segments having a resemblance factor above a particular threshold. Based upon the generated fingerprint, a data source may be filtered to block or to pass data corresponding to the generated fingerprint. The digital data may be audio data, video data, or other data.

What is claimed is: 
     
         1 . A method for isolating separate sources of audio within an audio signal, the method comprising:
 at a computing system, receiving an audio signal that includes audio from a plurality of subsets of audio, audio of the different subsets of the plurality of subsets having different source characteristics and originating from different sources;   separating the audio signal into a plurality of segments of audio data;   comparing the plurality of segments to one or more audio fingerprints; and   based on a comparison of the plurality of segments to the one or more audio fingerprints, separating at least one second segment of the plurality of segments corresponding to at least one second subset of the plurality of subsets of the audio data from at least one first segment of the plurality of segments corresponding to a first subset of the plurality of subsets of the audio signal.   
     
     
         2 . The method recited in  claim 1 , further comprising:
 outputting improved audio data that includes the at least one first segment, and substantially lacks the at least one second segment.   
     
     
         3 . The method recited in  claim 2 , wherein outputting the improved audio data includes:
 outputting an audio stream that includes of the improved audio data.   
     
     
         4 . The method recited in  claim 1 , wherein separating the at least one second segment from the first segment includes applying one or more masks or filters to the plurality of segments to obtain improved audio data that includes audio of the first segment substantially independent of audio of the at least one second segment. 
     
     
         5 . The method recited in  claim 1 , wherein receiving the audio signal includes receiving audio data in real time. 
     
     
         6 . The method recited in  claim 5 , wherein receiving the audio signal includes receiving data from a live microphone. 
     
     
         7 . The method recited in  claim 1 , wherein the at least one second segment comprises noise. 
     
     
         8 . The method recited in  claim 1 , wherein separating the audio signal into the plurality of segments includes identifying at least one segment by a frequency progression. 
     
     
         9 . The method recited in  claim 1 , wherein separating the audio signal into the plurality of segments includes identifying at least one segment as a continuous deviation above a baseline. 
     
     
         10 . The method recited in  claim 1 , wherein receiving the audio signal comprises receiving an audio signal from a combination of any of:
 a first voice;   a second voice;   a first musical instrument;   a second musical instrument; and   background noise.   
     
     
         11 . A computer-readable medium storing a computer program for performing a method for isolating subsets of audio data from different sources from an audio signal, the computer-readable medium comprising:
 computer storage media; and   computer-executable instructions stored on the computer storage media, which computer-executable instructions, when executed by a computing system, are configured to cause the computing system to:
 access an audio signal, the audio signal collectively including a plurality of subsets of audio, audio from different subsets of the plurality of subsets having different source characteristics and originating from different sources; 
 separate the audio signal into a plurality of segments of audio data; 
 compare the plurality of segments to one or more audio fingerprints and determine a resemblance value between each segment of the plurality of segments and at least one corresponding audio fingerprint of the one or more audio fingerprints; 
 compare each resemblance value above a threshold to determine whether the segment corresponding to that resemblance value is likely to correspond to a first subset of the plurality of subsets; and 
 output improved audio data, the improved audio data including an assembly of segments of the plurality of segments determined to likely correspond to the first subset, and the improved audio data substantially lacking segments determined to not likely correspond to the first subset. 
   
     
     
         12 . The computer-readable medium recited in  claim 11 , further comprising:
 the one or more audio fingerprints stored on the computer storage media.   
     
     
         13 . The computer-readable medium recited in  claim 11 , wherein the computer-executable instructions are configured to cause the computing system to output the improved audio data including segments determined to likely correspond to the first subset and substantially lacking noise. 
     
     
         14 . The computer-readable medium recited in  claim 11 , wherein the computer-executable instructions are configured to cause the computing system to output the improved audio data including segments determined to likely correspond to the first subset and substantially lacking segments determined likely to correspond to a particular voice. 
     
     
         15 . The computer-readable medium recited in  claim 11 , wherein the plurality of segments includes a combination of notes, progressions, or syllables. 
     
     
         16 . A computing system for isolating one or more samples within an audio signal, comprising:
 one or more processors;   data storage communicatively coupled to the one or more processors; and   computer-executable instructions that, when executed by the one or more processors, cause the computing system to:
 receive audio data over a network, the audio data being received by the network from an electronic device; 
 slice the audio data into a plurality of segments of audio data; 
 compare at least some segments of the plurality of segments to one or more audio fingerprints associated with a particular subset of a source of audio; 
 based on the comparison of the at least some segments to the one or more audio fingerprints, classify a subset of the at least some segments as segments likely originating from the particular subset of the source; and 
 separate the segments likely originating from the particular subset of the source from others of the at least some segments not likely originating from the particular subset of the source; 
 assembly improved audio data based on the segments likely originating from the particular subset of the source; and 
 output the improved audio data. 
   
     
     
         17 . The computing system of  claim 16 , wherein the computing system comprises a portable electronic device. 
     
     
         18 . The computing system of  claim 16 , wherein the computing system comprises a mobile telephone. 
     
     
         19 . A method for assembling a partial audio signal, comprising:
 receiving an audio signal, the audio signal including combined subsets of audio from two or more sources;   separating the audio signal into multiple segments of audio data, each segment of the multiple segments having a starting time and an ending time;   obtaining a resemblance value for each segment of the multiple segments relative to a plurality of fingerprints of audio data associated with a first subset of audio of the combined subsets of audio from two or more sources;   forming a first set of segments of the multiple segments, the first set including all segments of the multiple segments having resemblance values above a predetermined threshold with a fingerprint of the plurality of fingerprints of audio data associated with the first subset;   forming a second set of segments of the multiple segments, the second set including at least some segments of the multiple segments which are not in the first set; and   outputting an assembled audio signal, the assembled audio signal including the segments of the first set and substantially excluding segments of the second set.   
     
     
         20 . The method recited in  claim 19 , wherein receiving the audio signal includes accessing audio data. 
     
     
         21 . The method recited in  claim 20 , wherein receiving the audio signal includes receiving an audio signal including two or more subsets of audio from separate sources. 
     
     
         22 . The method recited in  claim 20 , wherein forming the second set of segments includes including segments corresponding to noise in the second set. 
     
     
         23 . The method recited in  claim 20 , further comprising:
 classifying the first set of segments as likely originating from a first subset of audio; and   classifying the second set of segments as likely originating from at least one second subset of audio.