Real time popularity based audible content aquisition
View Patent ↗A personalized news service provides personalized news programs for its users by generating personalized combinations of audible versions of news stories derived from text-based based versions of the news stories. The audible versions may be generated from the text-based version by a text-to-speech system, or may by recording a person reading aloud the text-based version. To acquire recordings, the personalized news service can make a determination that a particular news story has a threshold extent of popularity. The news service can then transmit a request to a remote recording station for a recording of a verbal reading of the particular news story. The news service can then receive the requested recording from the remote recording station.
1. A method comprising:
analyzing, by a computing device, generated playlists for a plurality of users, wherein the generated playlists contain news stories personalized for the users, and wherein text-based versions of at least some of the news stories are associated with audible versions automatically generated by a computerized text-to-speech (TTS) system;
making, by the computing device, a determination (i) that a particular news story in the generated playlists has at least a threshold extent of popularity, and (ii) that a text-based version of the particular news story is associated with an audible version of the particular news story that was generated by the TTS system;
based on making the determination, selecting, by the computing device, a particular remote news studio from a plurality of remote news studios;
transmitting, to the selected remote news studio, a request for a recording of a verbal reading of the text-based version of the particular news story;
receiving, from the selected remote news studio, the requested recording of the verbal reading of the text-based version of the particular news story;
updating, by the computing device, the generated playlists to associate the verbally-read version of the particular news story, instead of the audible version of the particular news story that was generated by the TTS system, with the text-based version of the particular news story; and
transmitting, by the computing device, a particular generated playlist of the generated playlists to a client device, wherein reception of the particular generated playlist by the client device triggers the client device to: (i) traverse the particular generated playlist, (ii) retrieve audio file representations of the audible versions of the news stories in the particular generated playlist, and (iii) play out the audio file representations.
2. The method of claim 1 , further comprising:
receiving, from the client device, a request for data representing an audible version of the text-based version of the particular news story; and
if the request is received prior to receiving the requested recording, fulfilling the received request by transmitting, to the client device, the audible version of the particular news story that was generated by the TTS system, or rather if the request is received subsequent to receiving the requested recording, fulfilling the received request by transmitting, to the client device, the verbally-read version of the particular news story.
3. The method of claim 1 , further comprising:
maintaining in data storage a first digital audio file representing the audible version of the particular news story that was generated by the TTS system, wherein the first digital audio file was generated based on a text file representing the text-based version of the particular news story, and wherein the text file is associated with the first digital audio file by a lookup reference; and
responsive to receiving the requested recording, (i) storing a second digital audio file of the verbally-read version of the particular news story in data storage, and (ii) updating the lookup reference to refer to the second digital audio file.
4. The method of claim 1 , wherein the determination is a first determination, wherein making the first determination comprises making a second determination that the audible version of the particular news story that was generated by the TTS system has at least the threshold extent of popularity.
5. The method of claim 4 , wherein making the second determination comprises (i) determining an extent to which the audible version of the particular news story that was generated by the TTS system has been selected in connection with generating playlists for personalized news programs for multiple users, and (ii) determining that the determined extent of selection exceeds a threshold extent of selection.
6. The method of claim 5 , further comprising:
generating the playlists, wherein generating the playlists comprises selecting, for each given one of the multiple users, a personalized combination of audible versions of news stories that were generated by the TTS system based, at least in part, on a correspondence between (i) an attribute associated with the selected audible versions of news stories that were generated by the TTS system, and (ii) an attribute associated with the given user.
7. The method of claim 6 ,
wherein selecting, for each given one of the multiple users, the personalized combination of audible versions of news stories that were generated by the TTS system comprises: (i) assigning rankings to each one of multiple audible versions of news stories that were generated by the TTS system based, at least in part, on an extent of correspondence between attributes associated with the audible versions of news stories that were generated by the TTS system and attributes associated with the given user, and (ii) selecting, from the multiple audible versions of news stories that were generated by the TTS system, a set of audible versions of news stories that were generated by the TTS system with the highest assigned rankings for inclusion in the personalized combination of audible versions of news stories that were generated by the TTS system for the given user.
8. The method of claim 5 ,
wherein determining the extent to which the audible version of the particular news story that was generated by the TTS system has been selected in connection with generating playlists for personalized news programs comprises determining a value representing the extent to which the audible version of the particular news story that was generated by the TTS system has been selected in connection with generating the playlists, and wherein determining that the determined extent of selection exceeds a threshold extent of selection comprises determining that the determined value exceeds a threshold value.
9. The method of claim 8 ,
wherein determining the value comprises determining the value based, at least in part, on a frequency at which the audible version of the particular news story that was generated by the TTS system is selected in connection with generating the playlists.
10. The method of claim 8 ,
wherein determining the value comprises determining the value based, at least in part, on a number of instances of the audible version of the particular news story that was generated by the TTS system being selected in connection with generating the playlists.
11. The method of claim 1 , wherein selecting the particular remote news studio from the plurality of remote news studios comprises selecting the particular remote news studio based on a correspondence between a geographic location of the particular remote news studios and an attribute associated with the particular news story.
12. The method of claim 1 , wherein selecting the particular remote news studio from the plurality of remote news studios comprises selecting the particular remote news studio based on an availability of a person affiliated with the particular remote news studio during an upcoming time interval.
13. The method of claim 1 , wherein selecting the particular remote news studio from the plurality of remote news studios comprises:
determining respective values representing respective extents of popularity of the respective news stories of the news stories in the generated playlists;
generating a ranked list of a group of the respective news stories having threshold extents of popularity, wherein the ranked list is ordered according to the respective values representing the respective extents of popularity in the group; and
wherein transmitting the request comprises transmitting an indication of the ranked list to the plurality of remote news studios.
14. The method of claim 13 , wherein selecting the particular remote news studio from the plurality of remote news studios comprises:
receiving, from one of the plurality of remote news studios, a communication that indicates a selection of one of the news stories in the group; and
updating the ranked list by removing the selected one of the news stories in the group and re-ordering the remaining news stories in the group according to the respective values representing the respective extents of popularity of the remaining news stories in the group; and
transmitting the updated ranked list to the plurality of remote news studios.
15. A computing system comprising:
a processor;
a memory;
a communication interface; and
program instructions, stored in the memory, that when executed by the processor, cause the computing system to perform operations comprising:
analyzing generated playlists for a plurality of users, wherein the generated playlists contain news stories personalized for the users, and wherein text-based versions of at least some of the news stories are associated with audible versions automatically generated by a computerized text-to-speech (TTS) system;
making a determination (i) that a particular news story in the generated playlists has at least a threshold extent of popularity, and (ii) that a text-based version of the particular news story is associated with an audible version of the particular news story that was generated by the TTS system;
based on making the determination, selecting a particular remote news studio from a plurality of remote news studios;
transmitting, to the selected remote news studio, via the communication interface, a request for a recording of a verbal reading of the text-based version of the particular news story;
receiving, via the communication interface, the requested recording of the verbal reading of the text-based version of the particular news story;
updating the generated playlists to associate the verbally-read version of the particular news story, instead of the audible version of the particular news story that was generated by the TTS system, with the text-based version of the particular news story; and
transmitting, via the communication interface, a particular generated playlist of the generated playlists to a client device, wherein reception of the particular generated playlist by the client device triggers the client device to: (i) traverse the particular generated playlist, (ii) retrieve audio file representations of the audible versions of the news stories in the particular generated playlist, and (iii) play out the audio file representations.
16. The computing system of claim 15 , wherein the program instructions cause the computing system to perform further operations comprising:
receiving, from the client device, a request for data representing an audible version of the text-based version of the particular news story; and
if the request is received prior to receiving the requested recording, transmitting, to the client device, the audible version of the particular news story that was generated by the TTS system, or rather if the request is received subsequent to receiving the requested recording, transmitting, to the client device, the verbally-read version of the particular news story.
17. The computing system of claim 15 , wherein the program instructions cause the computing system to perform further operations comprising:
maintaining in the memory a first digital audio file representing the audible version of the particular news story that was generated by the TTS system, wherein the first digital audio file was generated based on a text file representing the text-based version of the particular news story, and wherein the text file is associated with the first digital audio file by a lookup reference; and
responsive to receiving the requested recording, storing a second digital audio file of the verbally-read version of the particular news story in the memory, and updating the lookup reference to refer to the second digital audio file.
18. The computing system of claim 15 , wherein selecting the particular remote news studios from the plurality of remote news studios comprises selecting the particular remote news studio based on one or more of (a) a correspondence between a geographic location of the particular remote news studio and an attribute associated with the particular news story, or (b) an availability of a person affiliated with the particular remote news studio during an upcoming time interval.
19. A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause performance of operations comprising:
analyzing generated playlists for a plurality of users, wherein the generated playlists contain news stories personalized for the users, and wherein text-based versions of at least some of the news stories are associated with audible versions automatically generated by a computerized text-to-speech (TTS) system;
making a determination (i) that a particular news story in the generated playlists has at least a threshold extent of popularity, and (ii) that a text-based version of the particular news story is associated with an audible version of the particular news story that was generated by the TTS system;
based on making the determination, selecting a particular remote news studio from a plurality of remote news studios;
transmitting, to the selected remote news studio, a request for a recording of a verbal reading of the text-based version of the particular news story;
receiving the requested recording of the verbal reading of the text-based version of the particular news story;
updating the generated playlists to associate the verbally-read version of the particular news story, instead of the audible version of the particular news story that was generated by the TTS system, with the text-based version of the particular news story; and
transmitting a particular generated playlist of the generated playlists to a client device, wherein reception of the particular generated playlist by the client device triggers the client device to: (i) traverse the particular generated playlist, (ii) retrieve audio file representations of the audible versions of the news stories in the particular generated playlist, and (iii) play out the audio file representations.
20. The non-transitory computer-readable medium of claim 19 , wherein the instructions further cause performance of operations comprising:
receiving, from the client device, a request for data representing an audible version of the text-based version of the particular news story; and
if the request is received prior to receiving the requested recording, transmitting, to the client device, data representing the audible version of the particular news story that was generated by the TTS system, or rather if the request is received subsequent to receiving the requested recording, transmitting, to the client device, the verbally-read version of the particular news story.