Brian C Whitman

US Patent:

7772478, Aug 10, 2010

Filed:

Apr 12, 2007

Appl. No.:

11/734740

Inventors:

Brian A. Whitman - Somerville MA, US
Barry Vercoe - Natick MA, US

Assignee:

Massachusetts Institute of Technology - Cambridge MA

International Classification:

A63H 5/00

US Classification:

84609, 84600, 84603, 84615, 84616

Abstract:

There are disclosed methods and apparatus for understanding music. A classifier machine may be trained for each of a plurality of selected terms using a first plurality of music samples. The classifier machines may then be tested using a second plurality of music samples. The results from testing the classifier machines may then be used to select a plurality of semantic basis function from the selected terms. A semantic basis classifier machine may then be trained for each semantic basis function.

US Patent:

7949649, May 24, 2011

Filed:

Apr 10, 2008

Appl. No.:

12/100966

Inventors:

Brian Whitman - Somerville MA, US

Assignee:

The Echo Nest Corporation - Somerville MA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707710, 707802, 707913, 700 94

Abstract:

There are disclosed methods, apparatus, and storage media for building a music metadata library. Acoustic metadata for one or more music tracks may be automatically acquired from a plurality of client computing devices. Cultural metadata for the one or more music tracks may be automatically acquired by searching Internet-accessible content. The acquired acoustic and cultural metadata may be stored in the music metadata library.

US Patent:

8073854, Dec 6, 2011

Filed:

Apr 10, 2008

Appl. No.:

12/101013

Inventors:

Brian Whitman - Somerville MA, US
Tristan Jehan - Somerville MA, US

Assignee:

The Echo Nest Corporation - Somerville MA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707749, 707802, 707916

Abstract:

There is disclosed a method and storage media for determining the similarity between a first music collection and a second music collection. The similarity may be determined by comparing metadata associated with the first music collection and metadata associated with the second music collection. A plurality of comparison algorithms may be used including at least one of a comparison algorithm based on acoustic metadata and a comparison algorithm based on cultural metadata. The result of the plurality of comparison algorithms may be combined.

US Patent:

8280889, Oct 2, 2012

Filed:

May 19, 2011

Appl. No.:

13/111819

Inventors:

Brian Whitman - Cambridge MA, US

Assignee:

The Echo Nest Corporation - Somerville MA

International Classification:

G06F 7/00
G06F 17/30

US Classification:

707741, 707802, 707913, 707916, 700 94

Abstract:

Methods, apparatus, and storage media to build a music metadata library from music collections stored on a plurality of client computing devices. A music server may receive, from a first client computing device of the plurality of client computing device, a unique identifier for a track stored in the first client computing device. The music server may determine if the track is a known track corresponding to acoustic metadata that is already contained in the music metadata library. When the track is not a known track, the music server may send a message indicating the track is a new track to the first client computing device. The music server may then receive acoustic metadata for the track from the first client computing device and store the received acoustic metadata in the music metadata library.

US Patent:

8492633, Jul 23, 2013

Filed:

Jun 12, 2012

Appl. No.:

13/494183

Inventors:

Brian Whitman - Cambridge MA, US
Andrew Nesbit - London, GB
Daniel Ellis - New York NY, US

Assignee:

The Echo Nest Corporation - Summerville MA

International Classification:

A63H 5/00
G04B 13/00
G10H 7/00

US Classification:

84609

Abstract:

A method for fingerprinting an unknown music sample is disclosed. A plurality of known tracks may be segmented into reference samples. A reference fingerprint including a plurality of codes may be generated for each reference sample. An inverted index including, for each possible code value, a list of reference samples having reference fingerprints that contain the respective code value may be generated. An unknown fingerprint including a plurality of codes may be generated from the unknown music sample. A code match histogram may list candidate reference samples and associated scores, each score indicating a number of codes from the unknown fingerprint that match codes in the reference fingerprint. Time difference histograms may be generated for two or more reference samples having the highest scores. A determination may be made whether or not a single reference sample matches the unknown music sample based on a comparison of the time difference histograms.

US Patent:

20130262469, Oct 3, 2013

Filed:

Mar 22, 2013

Appl. No.:

13/849167

Inventors:

Brian Whitman - Cambridge MA, US

Assignee:

The Echo Nest Corporation - Somerville MA

International Classification:

G06F 17/30

US Classification:

707741

Abstract:

Methods, systems and computer program products are provided for predicting data. A name or title is obtained from a taste profile. There is an index into a data set based on the name or title, and a set of terms and corresponding term weights associated with the name or title are retrieved. A sparse vector is constructed based on the set of terms and term weights. The sparse vector is input to a training model including target data. The target data includes a subset of test data which has a correspondence to a predetermined target metric of data. A respective binary value and confidence level is output for each term, corresponding to an association between the term and the target metric.

US Patent:

20200159744, May 21, 2020

Filed:

Jan 22, 2020

Appl. No.:

16/749622

Inventors:

- Stockholm, SE
Brian Whitman - Cambridge MA, US

Assignee:

SPOTIFY AB - Stockholm

International Classification:

G06F 16/26
G06F 16/2457
G06F 16/735
G06F 16/635
G06F 16/435
G06F 16/68
G06N 20/00
G06N 7/00
G06N 5/00

Abstract:

Methods, systems and computer program products are provided for cross-media recommendation by store a plurality of taste profiles corresponding to a first domain and a plurality of media item vectors corresponding to a second domain. An evaluation taste profile in the first domain is applied to a plurality of models that have been generated based on relationship among the plurality of taste profiles and the plurality of media item vectors, and obtain a plurality of resulting codes corresponding to at least one of the plurality of media item vectors in the second domain.

US Patent:

20170139912, May 18, 2017

Filed:

Jan 31, 2017

Appl. No.:

15/420311

Inventors:

- Stockholm, SE
Brian Whitman - Cambridge MA, US

International Classification:

G06F 17/30
G06N 7/00
G06N 99/00

Abstract:

Methods, systems and computer program products are provided for cross-media recommendation by store a plurality of taste profiles corresponding to a first domain and a plurality of media item vectors corresponding to a second domain. An evaluation taste profile in the first domain is applied to a plurality of models that have been generated based on relationship among the plurality of taste profiles and the plurality of media item vectors, and obtain a plurality of resulting codes corresponding to at least one of the plurality of media item vectors in the second domain.