Back to results
US 9,934,785 B1Granted
Identification of taste attributes from an audio signal
AssigneeSpotify AB
InventorStéphane Hulaud
Life of this patent
Estimating the term…
Abstract
A system, method and computer product are provided for processing audio signals. An audio signal of a voice and background noise is input, and speech recognition is performed to retrieve speech content of the voice. There is retrieval of content metadata corresponding to the speech content, and environmental metadata corresponding to the background noise. There is a determination of preferences for media content corresponding to the content metadata and the environmental metadata, and an output is provided corresponding to the preferences.
Claims19
A method of processing audio signals, comprising the steps of:
- retrieving at least one content object generated by an audio signal processing system, the at least one content object including (1) one or more text objects corresponding to one or more words and (2) one or more text object data elements representing (i) a type representing the one or more text objects as at least one of a sentence, a verb, a noun phrase a determiner, and an adjective (ii) an emphasis level value representing a level of emphasis of the one or more words, correspondingly and (iii) an electronically generated confidence level value representing a confidence level that the audio signal processing system has properly identified the one or more words from an audio signal;
- retrieving content metadata generated by the audio signal processing system, the content metadata corresponding to the one or more words and including (i) one or more content metadata objects and (ii) a content metadata object confidence level value representing the confidence level that the audio signal processing system has properly identified the one or more content metadata objects, wherein the one or more content metadata objects include at least one of an emotion object, a gender object, an age object and an accent object;
- retrieving environmental metadata, generated by the audio signal processing system, the environmental metadata corresponding to the background noise and including (i) one or more environmental metadata objects and (ii) an environmental metadata object confidence level value representing the audio signal processing system has properly identified the one or more environmental metadata objects from the audio signal;
- retrieving, from a profile database, a user profile containing historical listening practices of a user;
- determining, based on the user profile, the at least one content object, the content metadata and the environmental metadata, at least one of (i) media content and (ii) a recommended next media track;
- and
- outputting, on an electronic device, at least one of the media content and the recommended next media track, wherein the outputting comprises playing the media content or displaying the recommended next media track, correspondingly,
- wherein when the recommended next media track is selected, media content corresponding to the recommended next media track is played.
5 more elements
A system for processing audio signals, comprising:
- at least one processor configured to:
- retrieve at least one content object generated by an audio signal processing system, the at least one content object including (1) one or more text objects corresponding to one or more words and (2) one or more text object data elements representing (i) a type representing the one or more text objects as at least one of a sentence, a verb, a noun phrase a determiner, and an adjective (ii) an emphasis level value representing a level of emphasis of the one or more words, correspondingly and (iii) an electronically generated confidence level value representing a confidence level that the audio signal processing system has properly identified the one or more words from an audio signal;
- retrieve content metadata, generated by the audio signal processing system, the content metadata corresponding to the one or more words and including (i) one or more content metadata objects and (ii) a content metadata object confidence level value representing the confidence level that the audio signal processing system has properly identified the one or more content metadata objects, wherein the one or more content metadata objects include at least one of an emotion object, a gender object, an age object and an accent object;
- retrieve environmental metadata, generated by the audio signal processing system, the environmental metadata corresponding to the background noise and including (i) one or more environmental metadata objects and (ii) an environmental metadata object confidence level value representing the audio signal processing system has properly identified the one or more environmental metadata objects from the audio signal;
- retrieve, from a profile database, a user profile containing historical listening practices of a user;
- determine, based on the user profile, the at least one content object, the content metadata and the environmental metadata, at least one of (i) media content and (ii) a recommended next media track;
- and
- cause at least one of the media content and the recommended next media track to be output on an electronic device, wherein the outputting comprises playing the media content or displaying the recommended next media track, correspondingly,
- wherein when the recommended next media track is selected, media content corresponding to the recommended next media track is played.
6 more elements
A non-transitory computer-readable storage medium storing a program for causing a computer to implement a method of processing audio signals, comprising the steps of:
- retrieving at least one content object generated by an audio signal processing system, the at least one content object including (1) one or more text objects corresponding to one or more words and (2) one or more text object data elements representing at least one of (i) a type representing the one or more text objects as at least one of a sentence, a verb, a noun phrase a determiner, and an adjective (ii) an emphasis level value representing a level of emphasis of the one or more words, correspondingly and (iii) an electronically generated confidence level value representing a confidence level that the audio signal processing system has properly identified the one or more words from an audio signal;
- retrieving content metadata generated by the audio signal processing system, the content metadata corresponding to the one or more words and including (i) one or more content metadata objects and (ii) a content metadata object confidence level value representing the confidence level that the audio signal processing system has properly identified the one or more content metadata objects, wherein the one or more content metadata objects include at least one of an emotion object, a gender object, an age object and an accent object;
- retrieving environmental metadata, generated by the audio signal processing system, the environmental metadata corresponding to the background noise and including (i) one or more environmental metadata objects and (ii) an environmental metadata object confidence level value representing the audio signal processing system has properly identified the one or more environmental metadata objects from the audio signal;
- retrieving, from a profile database, a user profile containing historical listening practices of a user;
- determining, based on the user profile, the at least one content object, the content metadata and the environmental metadata, at least one of (i) media content and (ii) a recommended next media track;
- and
- outputting, on an electronic device, at least one of the media content and the recommended next media track, wherein the outputting comprises playing the media content or displaying the recommended next media track, correspondingly,
- wherein when the recommended next media track is selected, media content corresponding to the recommended next media track is played.
5 more elements
Dependencies are inferred from the claim text.