Daniel M Ellis

US Patent:

8135221, Mar 13, 2012

Filed:

Oct 7, 2009

Appl. No.:

12/574716

Inventors:

Wei Jiang - New York NY, US
Courtenay Cotton - New York NY, US
Shih-Fu Chang - New York NY, US
Daniel P. Ellis - New York NY, US
Alexander C. Loui - Penfield NY, US

Assignee:

Eastman Kodak Company - Rochester NY
Columbia University - Rochester NY

International Classification:

G06K 9/62
G06K 9/54

US Classification:

382224, 382305, 725 37

Abstract:

A method for determining a classification for a video segment, comprising the steps of: breaking the video segment into a plurality of short-term video slices, each including a plurality of video frames and an audio signal; analyzing the video frames for each short-term video slice to form a plurality of region tracks; analyzing each region track to form a visual feature vector and a motion feature vector; analyzing the audio signal for each short-term video slice to determine an audio feature vector; forming a plurality of short-term audio-visual atoms for each short-term video slice by combining the visual feature vector and the motion feature vector for a particular region track with the corresponding audio feature vector; and using a classifier to determine a classification for the video segment responsive to the short-term audio-visual atoms.

US Patent:

8586847, Nov 19, 2013

Filed:

Dec 2, 2011

Appl. No.:

13/310190

Inventors:

Daniel Ellis - New York NY, US
Brian Whitman - Cambridge MA, US

Assignee:

The Echo Nest Corporation - Summerville MA

International Classification:

G10H 3/00

US Classification:

84603, 700 94

Abstract:

Methods, computing devices, and machine readable storage media for generating a fingerprint of a music sample. The music sample may be filtered into a plurality of frequency bands. Onsets in each of the frequency bands may be independently detected. Inter-onset intervals between pairs of onsets within the same frequency band may be determined. At least one code associated with each onset may be generated, each code comprising a frequency band identifier identifying a frequency band in which the associated onset occurred and one or more inter-onset intervals. Each code may be associated with a timestamp indicating when the associated onset occurred within the music sample. All generated codes and the associated timestamps may be combined to form the fingerprint.

US Patent:

20080072741, Mar 27, 2008

Filed:

Sep 27, 2007

Appl. No.:

11/863014

Inventors:

Daniel Ellis - New York NY, US

International Classification:

G10H 7/00

US Classification:

084609000

Abstract:

Methods and systems for identifying similar songs are provided. In accordance with some embodiments, methods for identifying similar songs are provided, the methods comprising: identifying beats in at least a portion of a song; generating beat-level descriptors of the at least a portion of the song corresponding to the beats; comparing the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs. In accordance with some embodiments, systems for identifying similar songs are provided, the systems comprising: a digital processing device that: identifies beats in at least a portion of a song; generates beat-level descriptors of the at least a portion of the song corresponding to the beats; and compares the beat-level descriptors to other beat-level descriptors corresponding to a plurality of songs.

US Patent:

20090271182, Oct 29, 2009

Filed:

Feb 12, 2009

Appl. No.:

12/370424

Inventors:

Marios Athineos - New York NY, US
Daniel P.W. Ellis - New York NY, US

Assignee:

THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK - New York NY

International Classification:

G10L 21/06

US Classification:

704205, 704E21019

Abstract:

In accordance with the present invention, computer implemented methods and systems are provided for representing and modeling the temporal structure of audio signals. In response to receiving a signal, a time-to-frequency domain transformation on at least a portion of the received signal to generate a frequency domain representation is performed. The time-to-frequency domain transformation converts the signal from a time domain representation to the frequency domain representation. A frequency domain linear prediction (FDLP) is performed on the frequency domain representation to estimate a temporal envelope of the frequency domain representation. Based on the temporal envelope, one or more speech features are generated.

US Patent:

20110087349, Apr 14, 2011

Filed:

Oct 12, 2010

Appl. No.:

12/902859

Inventors:

Daniel P.W. Ellis - New York NY, US
Courtenay V. Cotton - New York NY, US

Assignee:

The Trustees of Columbia University in the City of New York - New York NY

International Classification:

G06F 17/00

US Classification:

700 94

Abstract:

System, methods, and media that: receive a first piece of audio content; identify a first plurality of atoms that describe at least a portion of the first piece of audio content using a Matching Pursuit algorithm; form a first group of atoms from at least a portion of the first plurality of atoms, the first group of atoms having first group parameters; form at least one first hash value for the first group of atoms based on the first group parameters; compare the at least one first hash value with at least one second hash value, wherein the at least one second hash value is based on second group parameters of a second group of atoms associated with a second piece of audio content; and identify a match between the first piece of audio content and the second piece of audio content based on the comparing.

US Patent:

20110116551, May 19, 2011

Filed:

Sep 13, 2010

Appl. No.:

12/880858

Inventors:

Aaron KLEIN - Flushing NY, US
Yannis TSIVIDIS - New York NY, US
Daniel P.W. ELLIS - New York NY, US

Assignee:

THE TRUSTEES OF COLUMBIA UNIVERSITY IN THE CITY OF NEW YORK - New York NY

International Classification:

H04B 1/66
H04N 7/26

US Classification:

37524025, 375240, 37524001, 375E07027, 375E07026

Abstract:

Apparatus and methods for processing compression encoded signals are provided. In some embodiments, a signal processing method is provided that includes receiving a subband of a compression encoded signal at a subband processor, generating envelope information regarding the subband of the compression encoded signal to provide changes in the dynamic range of the compression encoded signal for fixed-point digital signal processing, processing the compression encoded signal with a fixed-point companding digital signal processor using the envelope information, and producing a processed compression encoded signal at the output of the subband processor.

US Patent:

20130091167, Apr 11, 2013

Filed:

Oct 5, 2012

Appl. No.:

13/646580

Inventors:

The Trustees of Columbia University in the City of - New York NY, US
Daniel P.W. Ellis - New York NY, US

Assignee:

The Trustees of Columbia University in the City of New York - New York NY

International Classification:

G06F 17/30

US Classification:

707769, 707E17014

Abstract:

Methods, systems, and media for identifying similar songs using jumpcodes are provided. In some embodiments, methods for a cover song from a query song are provided, the methods comprising: identifying a query song jumpcode for the query song, wherein the query song jumpcode is indicative of changes in prominent pitch over a portion of the query song; identifying a plurality of reference song jumpcodes for a reference song, wherein each of the reference song jumpcodes is indicative of changes in prominent pitch over a portion of the reference song; determining if the query song jumpcode matches any of the plurality of reference song jumpcodes; and upon determining that the query song jumpcode matches at least one of the plurality of reference song jumpcodes, generating an indication that the reference song is a cover song of the query song.

US Patent:

20130159756, Jun 20, 2013

Filed:

Mar 17, 2011

Appl. No.:

13/634568

Inventors:

Daniel P.W. Ellis - New York NY, US

International Classification:

G06F 1/28

US Classification:

713340

Abstract:

Methods and systems for analyzing gross resource consumption data to determine resource consumption of individual devices are disclosed. In some embodiments, the methods and methods include the following: obtaining a first data that includes a time series of a gross resource consumption for a location; using blind signal separation techniques, identifying power-on and power-off events within the first data, the events being caused by turning particular devices at the location on and off during the time series and each of the events reflecting a power consumption signature for each of the particular devices; associating each of the events with a known device, the known device being substantially similar to one of the particular devices at the location; and determining a portion of the gross resource consumption consumed by each of the particular devices.