Systems, methods, and apparatus to improve media identification

A computer-implemented method comprising:

• generating hashed media data values, wherein generating the hashed media data values comprises performing a hash function on reference audio data values;
• generating a first candidate list, wherein generating the first candidate list comprises mapping the hashed media data values to a first set of buckets in a hash table, wherein the first set of buckets comprises a first bucket size;
• identifying a first set of reference matches from the first candidate list, wherein the first set of reference matches satisfy a first threshold of hits;
• generating a second candidate list, wherein generating the second candidate list comprises mapping a subset of the first set of reference matches that do not satisfy the first threshold of hits to a second set of buckets in the hash table, wherein the second set of buckets comprises a second bucket size;
• identifying a second set of reference matches from the second candidate list, wherein the second set of reference matches satisfy a second threshold of hits;
• and
• identifying one or more candidate matches based on the identified first set of reference matches and the second set of reference matches, wherein the identifying one or more candidate matches comprises determining a number of times that a media ID appears in the identified first set of reference matches and the identified second set of reference matches.

A computing device comprising:

• one or more processors;
• and
• a tangible, non-transitory computer readable storage medium comprising instructions that, when executed, cause the one or more processors to perform a set of operations comprising:
• generating hashed media data values, wherein generating the hashed media data values comprises performing a hash function on reference audio data values;
• generating a first candidate list, wherein generating the first candidate list comprises mapping the hashed media data values to a first set of buckets in a hash table, wherein the first set of buckets comprises a first bucket size;
• identifying a first set of reference matches from the first candidate list, wherein the first set of reference matches satisfy a first threshold of hits;
• generating a second candidate list, wherein generating the second candidate list comprises mapping a subset of the first set of reference matches that do not satisfy the first threshold of hits to a second set of buckets in the hash table, wherein the second set of buckets comprises a second bucket size;
• identifying a second set of reference matches from the second candidate list, wherein the second set of reference matches satisfy a second threshold of hits;
• and
• identifying one or more candidate matches based on the identified first set of reference matches and the second set of reference matches, wherein the identifying one or more candidate matches comprises determining a number of times that a media ID appears in the identified first set of reference matches and the identified second set of reference matches.