Hold back and real time ranking of results in a streaming matching system

Patent metadata

Field	Value
Publication number	US-9754026-B2
Application number	US-201615348776-A
Country	US
Kind code	B2
Filing date	Nov 10, 2016
Priority date	Dec 31, 2012
Publication date	Sep 5, 2017
Grant date	Sep 5, 2017

Abstract

A matching system receives probe audio samples for comparison to references of a data store. Comparisons are generated between a first segment of a probe audio sample and corresponding time segments of a plurality of reference audio samples to identify a plurality of sufficiently matching reference audio samples based upon a first set of consistency scores. Matching references are retained, unless they meet a score threshold. Comparisons are continually generated with a second segment of the probe audio sample and corresponding time segments of the sufficiently matching reference audio samples to generate a second set of consistency scores. The retained results are outputted based on the first and second set of consistency scores.

First claim

What is claimed is:

1. A method, comprising:
using a processor to execute computer executable instructions stored on a non-transitory computer readable medium to perform operations comprising:
receiving, from an audio streaming system, a probe audio sample;
comparing a first time segment of the probe audio sample to corresponding time segments of a plurality of reference audio samples to identify a plurality of sufficiently matching reference audio samples based upon a first set of consistency scores generated between one or more feature vectors of the first time segment of the probe audio sample and corresponding feature vectors of the first time segment of each of the reference audio samples;
determining that the sufficiently matching reference audio samples do not meet a predetermined score threshold;
retaining the sufficiently matching reference audio samples;
comparing a second time segment of the probe audio sample to corresponding time segments of the sufficiently matching reference audio samples to identify a plurality of additional matching reference audio samples based upon a second set of consistency scores generated between one or more features vectors of the second time segment of the probe audio sample and corresponding feature vectors of the second time segment of each of the sufficiently matching reference audio samples; and
outputting at least one of the reference audio samples based on the first and second set of consistency scores.

2. The method of claim 1 , further comprising:
assigning respective ranking scores to the sufficiently matching reference audio samples based on the first set of consistency scores;
assigning respective ranking scores to the additional matching reference audio samples based on the second set of consistency scores;
separately weighting the first set of ranking scores and the second set of ranking scores; and
outputting at least one of the reference audio samples based on the weighted first and second set of ranking scores.

3. The method of claim 2 , further comprising outputting ones of the matching reference audio samples that have respective assigned ranking scores that exceed a defined ranking threshold.

4. The method of claim 3 , further comprising, after a defined amount of time of outputting ones of the matching reference audio samples that have respective ranking scores that exceed the defined ranking threshold, outputting the plurality of matching reference audio samples with respective assigned ranking scores.

5. The method of claim 4 , wherein the defined amount of time is based on an amount of the probe audio sample received.

6. The method of claim 1 , wherein the comparing the first time segment of the probe audio sample to the corresponding time segment of the reference audio samples to identify the at least one matching reference audio sample comprises:
generating a fingerprint of the first time segment of the probe audio sample; and
identifying respective fingerprints of the corresponding time segment of the reference audio samples that match the fingerprint of the first time segment of the probe audio sample by at least a defined matching threshold.

7. The method of claim 2 , wherein the respective ranking scores are generated based on at least one from the group consisting of:
respective availability measures indicating whether the respective matching reference audio samples are individually available for sale;
respective popularity measures indicating respective popularities of the matching reference audio samples;
respective compilation measures indicating whether the respective matching reference audio samples are from an original album or a compilation album; and
respective sales measures indicating respective sales amounts of the matching reference audio samples.

8. The method of claim 7 , wherein the respective ranking scores comprise a sequence of bit vectors comprising a set of components that each comprise a set of bits occupying most to least significant bit positions based on an importance of a corresponding measure.

9. The method of claim 7 , wherein the respective popularity measures are generated from heuristics associated with at least one parameter related to the respective matching reference audio samples.

10. A system, comprising:
a non-transitory memory that stores computer executable components; and
a microprocessor that executes the following computer executable components stored in the memory, the computer executable components comprising:
a receiving component that receives, from an audio streaming service, a probe audio sample;
a comparing component that compares a first time segment of the probe audio sample to corresponding time segments of a plurality of reference audio samples;
a matching component that identifies a plurality of sufficiently matching reference audio samples based upon a first set of consistency scores generated between one or more feature vectors of the first time segment of the probe audio sample and corresponding feature vectors of the first time segment of each of the reference audio samples;
a retaining component that determines that the sufficiently matching reference audio samples do not meet a predetermined score threshold and retains the sufficiently matching reference audio samples; and
wherein the comparing component is further configured to compare a second time segment of the probe audio sample to corresponding time segments of the sufficiently matching reference audio samples to identify a plurality of additional matching reference audio samples based upon a second set of consistency scores generated between one or more feature vectors of the second time segment of the probe audio sample and corresponding feature vectors of the second time segment of each of the sufficiently matching reference audio samples; and
a scoring component configured to output at least one of the reference audio samples based on the first and second set of consistency scores.

11. The system of claim 10 , further comprising:
a ranking component that assigns respective ranking scores to the sufficiently matching reference audio samples based on the first set of consistency scores;
wherein the ranking component is further configured to assign respective ranking scores to the additional matching reference audio samples based on the second set of consistency scores;
wherein the scoring component is further configured to separately weight the first set of ranking scores and the second set of ranking scores; and
wherein the scoring component is further configured to output at least one of the reference audio samples based on the weighted first and second set of ranking scores.

12. The system of claim 10 , wherein the second time segment includes the first time segment of the audio streaming content.

13. The system of claim 10 , wherein the scoring component dynamically determines which matching reference audio samples to output based on the score comparison.

14. The system of claim 10 , wherein the retaining component retains the matching reference audio samples based on a set of defined parameters that include at least one of a time for comparison operations, or an amount of the audio streaming content compared.

15. The system of claim 11 , wherein the respective ranking scores are generated based on at least one from the group consisting of: whether the respective matching reference audio samples are individually available for sale, respective popularities of the matching reference audio samples, whether the respective matching reference audio samples are from an original album or a compilatio

Assignees

• Google Inc

Inventors

• Roblek Dominik
• Sharifi Matthew

Classifications

• G10L25/54

  for retrieval · CPC title

• G06F17/30769

  Physics · mapped topic

• G06F17/30743 Primary

  Physics · mapped topic

• G06F17/30758

  Physics · mapped topic

• G06F16/638

  Presentation of query results · CPC title