Back to results
US 9,959,342 B2Granted
Audio augmented reality system
AssigneeMicrosoft Technology Licensing, LLC
InventorsNeal Osotio, Angela Moulden
Life of this patent
Estimating the term…
Abstract
Techniques for online information search and retrieval for a query including a digital audio waveform. In an aspect, an audio waveform is received and digitized by at least one of a plurality of audio input devices. The digitized audio waveforms are transmitted to a central processing unit, which formulates and submits a query to an online engine. The formulated query may include the at least one digital audio waveform. The online engine retrieves one or more online results relevant to the formulated query. The online results may include one or more relevant visual results, and/or one or more relevant audio results. The retrieved results are served in real-time back to a user, via a device having audio output capability, and/or a device having visual data output capability.
Claims16
A method comprising:
- receiving, by a processor, a query comprising at least three digital sound waveforms including a first digital sound waveform from a first source and a second digital sound waveform from a second source;
- retrieving, by an online engine, at least one online result relevant to both the first and second digital sound waveforms;
- generating, by a synthesizing device, a synthesized sound corresponding to the at least one online result;
- and
- serving the generated synthesized sound in response to the received query;
- wherein each sound waveform of the at least three digital sound waveforms being received by a device having a distinct location, the retrieving the at least one online result comprising:
- calculating a position of a source object based on the at least three digital sound waveforms, the calculating comprising determining a relative delay amongst the at least three digital sound waveforms, and determining relative positions of the distinct locations of the devices based on the determined relative delay.
4 more elements
A method comprising:
- receiving, by a processor, a query comprising a first digital sound waveform from a first source and a second digital sound waveform from a second source;
- retrieving, by an online engine, at least one online result relevant to both the first and second digital sound waveforms;
- generating, by a synthesizing device, a synthesized sound corresponding to the at least one online result;
- serving the generated synthesized sound in response to the received query;
- receiving, by the online engine, an indication of user approval of the served generated synthesized sound;
- and based on the received indication of user approval, updating a machine learning algorithm used to retrieve the at least one online result relevant to the received query, the updating comprising training the machine learning algorithm using the retrieved at least one online result as a training vector.
3 more elements
An apparatus comprising:
- a query processing module configured to receive a query comprising at least three digital sound waveforms including a first digital sound waveform from a first source and a second digital sound waveform from a second source;
- a search engine configured to retrieve at least one online result relevant to both the first and second digital sound waveforms;
- a synthesis module configured to generate a synthesized sound corresponding to the at least one online result;
- and
- a transmission module configured to serve the generated synthesized sound in response to the received query;
- wherein each sound waveform of the at least three digital sound waveforms being received by a device having a distinct location, the search engine further configured to:
- calculate a position of a source object based on the at least three digital sound waveforms, the calculating comprising determining a relative delay amongst the at least three digital sound waveforms and determining relative positions of the distinct locations of the devices based on the determined relative delay;
- and retrieve the calculated position of the source object in response to the received query.
5 more elements
An apparatus comprising:
- a query processing module configured to receive a query comprising a first digital sound waveform from a first source and a second digital sound waveform from a second source;
- a search engine configured to retrieve at least one online result relevant to both the first and second digital sound waveforms;
- a synthesis module configured to generate a synthesized sound corresponding to the at least one online result;
- and
- a transmission module configured to serve the generated synthesized sound in response to the received query, the search engine further configured to receive an indication of user approval of the served generated synthesized sound;
- and based on the received indication of user approval, update a machine learning algorithm used to retrieve the at least one online result relevant to the received query, the updating comprising training the machine learning algorithm using the retrieved at least one online result as a training vector.
3 more elements
Dependencies are inferred from the claim text.