Digital voice communication advertising
First Claim
Patent Images
1. A computer-implemented method for processing audio data, the computer-implemented method comprising:
- receiving, by a recognition module in a server, an audio conversation from a client device;
converting, by the recognition module, the audio conversation into text that is representative of the audio conversation;
sampling, by the recognition module, the text for a word;
buffering, by the recognition module, at least a portion of the text during the sampling;
searching, by a content module in the server, in a database for a piece of content that is associated with the word, the searching including a backward sampling, wherein the backward sampling comprises;
identifying the word in the portion of the text, the word occurring at a first frequency that is less than a second frequency associated with a plurality of other words in conversations; and
at least partly in response to identifying the word, looking for a relevant word in a segment of the portion of the text, wherein the segment of the portion precedes the word, and wherein detection of the relevant word in combination with the word triggers provisioning of the piece of content; and
sending, by the content module, the provisioned piece of content to the client device.
2 Assignments
0 Petitions
Accused Products
Abstract
Techniques for providing third party incentivized communications sessions are described. In some embodiments, a Voice over IP (VoIP) conversation is sampled for audio information such as a word, phrase, utterance, or phoneme that may be associated with an advertisement in a database. The database may be searched for the audio information, or in some embodiments a text equivalent of the audio information, and the associated advertisement may be served to a participant in the VoIP communication session. In some implementations, data from or about the conversation may be compiled for analysis.
-
Citations
19 Claims
-
1. A computer-implemented method for processing audio data, the computer-implemented method comprising:
-
receiving, by a recognition module in a server, an audio conversation from a client device; converting, by the recognition module, the audio conversation into text that is representative of the audio conversation; sampling, by the recognition module, the text for a word; buffering, by the recognition module, at least a portion of the text during the sampling; searching, by a content module in the server, in a database for a piece of content that is associated with the word, the searching including a backward sampling, wherein the backward sampling comprises; identifying the word in the portion of the text, the word occurring at a first frequency that is less than a second frequency associated with a plurality of other words in conversations; and at least partly in response to identifying the word, looking for a relevant word in a segment of the portion of the text, wherein the segment of the portion precedes the word, and wherein detection of the relevant word in combination with the word triggers provisioning of the piece of content; and sending, by the content module, the provisioned piece of content to the client device. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. One or more computer-readable storage media comprising computer-executable instructions that, when executed, direct a computing system to:
-
receive a portion of audio content from a client device; convert the portion of the audio content into a computer recognizable representation of the portion of the audio content; buffer at least a portion of the computer recognizable representation of the portion of the audio content to generate a buffered portion of the computer recognizable representation of the portion of the audio content; identify a word in the buffered portion, the word occurring at a first frequency that is less than a second frequency associated with a plurality of other words in conversations; at least partly in response to identifying the word, identify a relevant word in a segment of the buffered portion that precedes the word; at least partly in response to identifying the relevant word, search a database for at least one piece of content that is associated with the word, the search locating the at least one piece of content; in response to the search locating the at least one piece of content, send the at least one piece of content to the client device; and assess a third party a fee for sending the at least one piece of content. - View Dependent Claims (12, 13, 14)
-
-
15. A system comprising:
-
one or more processors; memory, communicatively coupled to the one or more processors, that stores; a database including a record associating a pre-selected word with a piece of content, wherein the pre-selected word occurs at a first frequency that is less than a second frequency associated with a plurality of other words in conversations; one or more modules that; receive an audio session from a client device; convert the audio session to a computer recognizable representation of the audio session; identify, in a buffered portion of the computer recognizable representation of the audio session, a word that matches the pre-selected word; upon identifying the word that matches the pre-selected word, identify another word that precedes the word in the buffered portion of the computer recognizable representation of the audio session; and upon identifying the other word, send the piece of content to the client device based at least in part on one or more parameters. - View Dependent Claims (16, 17, 18, 19)
-
Specification