Methods and Apparatus for Advertising Using a Frequency of Spoken Word Database
An electronic device performs a method for advertising using a frequency of spoken word database. The method includes sending a plurality of observation phrases to a plurality of mobile devices and receiving, from a subset of the plurality of mobile devices for each observation phrase in a subset of the plurality of observation phrases, at least one utterance statistic that indicates how often the utterance of observation phrase was detected by the subset of the plurality of mobile devices over a time period. The method further includes building a frequency of spoken word database including the subset of the plurality of observation phrases each having associated therewith a set frequency of utterance values derived from the at least one utterance statistic received for that observation phrase. The frequency of spoken word database is used to determine at least one advertisement and linked observation phrase to send to a mobile device.
1 . A method performed by an electronic device for advertising using a frequency of spoken word database, the method comprising:
sending a plurality of observation phrases to a plurality of mobile devices;
receiving, from a subset of the plurality of mobile devices for each observation phrase in a subset of the plurality of observation phrases, at least one utterance statistic that indicates how often utterance of the observation phrase was detected over a time period;
building a frequency of spoken word database comprising the subset of the plurality of observation phrases each having associated therewith a set of frequency of utterance values derived from the at least one utterance statistic received for that observation phrase, wherein the frequency of spoken word database is used to determine at least one advertisement and linked observation phrase to send to a first mobile device.
2 . The method of claim 1 further comprising filtering the frequency of spoken word database by at least one filter parameter to determine the at least one advertisement and linked observation phrase to send to the first mobile device.
3 . The method of claim 2 , wherein the at least one filter parameter comprises at least one of:
time of day;
geographic region;
user characteristic data;
device characteristic data.
4 . The method of claim 2 further comprising:
receiving current device data from the first mobile device;
determining, based on the current device data, the at least one filter parameter for filtering the frequency of spoken word database to determine the at least one advertisement and linked observation phrase to send to the first mobile device.
5 . The method of claim 1 further comprising:
receiving from the first mobile device current device data;
determining a first set of advertisements that are linked to a first set of observation phrases from the frequency of spoken word database based on the current device data and based on the frequency of utterance values for the first set of observation phrases;
sending the first set of advertisements and linked first set of observation phrases to the first mobile device.
6 . The method of claim 5 , wherein each observation phrase in the linked first set of observation phrases is assigned an importance value.
7 . The method of claim 6 , wherein the first set of advertisements and linked first set of observation phrases is further determined based on the importance values for the first set of observation phrases.
8 . The method of claim 7 further comprising:
determining a candidate set of advertisements based on the current device data;
prioritizing the candidate set of advertisements to determine the first set of advertisements and linked first set of observation phrases based on the frequency of utterance values and importance values for the first set of observation phrases.
9 . The method of claim 8 , wherein the candidate set of advertisements is prioritized to determine the first set of advertisements and linked first set of observation phrases using a function comprising frequency of utterance value*importance value.
10 . The method of claim 6 , wherein the importance value is assigned to each observation phrase in the linked first set of observation phrases based on at least one of:
order of the observation phrase within a list;
numerical ranking of the observation phrase;
monetary bidding of the observation phrase.
11 . The method of claim 5 , wherein the current device data comprises at least one of:
device location data;
time of day;
device characteristic data;
user characteristic data;
user history of selected advertisements;
user search history.
12 . The method of claim 5 further comprising:
receiving, from the first mobile device, at least one updated utterance statistic for a first observation phrase in the linked first set of observation phrases;
updating, using the at least one updated utterance statistic, the set of frequency of utterance values for the first observation phrase in the frequency of spoken word database.
13 . The method of claim 5 , wherein a first advertisement in the first set of advertisements is linked to multiple observation phrases in the linked first set of observation phrases, the method further comprising indicating to the first mobile device a priority for detecting utterance of the multiple observation phrases in order to display the first advertisement.
14 . The method of claim 13 , wherein the priority for detecting the utterance of the multiple observation phrases is based on at least one of:
time of day;
geographic region.
15 . A system configured for advertising using a frequency of spoken word database, the system comprising:
a network interface configured to:
send a plurality of observation phrases to a plurality of mobile devices; and
receive, from a subset of the plurality of mobile devices for each observation phrase in a subset of the plurality of observation phrases, at least one utterance statistic that indicates how often utterance of the observation phrase was detected over a time period;
a processor coupled to the network interface and configured to:
build a frequency of spoken word database comprising the subset of the plurality of observation phrases each having associated therewith a set of frequency of utterance values derived from the at least one utterance statistic received for that observation phrase; and
determine, using the frequency of spoken word database, at least one advertisement and linked observation phrase to send to a first mobile device.
16 . The system of claim 15 , wherein:
the network interface is further configured to receive from the first mobile device current device data;
the processor is further configured to determine a first set of advertisements that are linked to a first set of observation phrases from the frequency of spoken word database based on the current device data and based on the frequency of utterance values for the first set of observation phrases;
the network interface is further configured to send the first set of advertisements and linked first set of observation phrases to the first mobile device.
17 . The system of claim 16 , wherein each observation phrase in the linked first set of observation phrases is assigned an importance value, wherein the processor is further configured to:
determine a candidate set of advertisements based on the current device data;
prioritize the candidate set of advertisements to determine the first set of advertisements and linked first set of observation phrases based on the frequency of utterance values and importance values for the first set of observation phrases.
18 . The system of claim 17 , wherein the processor is further configured to prioritize the set of advertisements to determine the first set of advertisements and linked first set of observation phrases using a function comprising frequency of utterance value*importance value.
19 . The system of claim 17 further comprising a memory component coupled to the processor, wherein the memory component has stored thereon:
the frequency of word database; and
an advertisement database that contains a plurality of advertisements each associated with at least one observation phrase that is assigned an importance value based on at least one of order of the observation phrase within a list, numerical ranking of the observation phrase, or monetary bidding of the observation phrase, wherein the frequency of spoken word database and the advertisement database are used to determine and prioritize the candidate set of advertisements.
20 . The system of claim 16 wherein:
the network interface is further configured to receive, from the first mobile device, at least one updated utterance statistic for a first observation phrase in the linked first set of observation phrases; and
the processor is further configured to update, using the at least one updated utterance statistic, the set of frequency of utterance values for the first observation phrase in the frequency of spoken word database.