User interface for streaming spoken query
View Patent ↗Methods and systems for information retrieval include analyzing audio data to produce word hypotheses. Displaying the word hypotheses in motion at different respective speeds at once across a graphical display. Information is retrieved in accordance with one or more selected terms from the displayed word hypotheses.
1. A method for information retrieval, comprising:
analyzing audio data using a processor to produce a plurality of word hypotheses;
displaying the plurality of word hypotheses in motion at different respective speeds at once across a graphical display, the speeds being based on a confidence value associated with each respective word hypothesis; and
retrieving in accordance with one or more selected terms from the displayed word hypotheses.
2. The method of claim 1 , wherein each of the plurality of word hypotheses has an associated importance value.
3. The method of claim 2 , wherein only word hypotheses having an importance above an importance threshold are displayed.
4. The method of claim 2 , wherein displaying the plurality of word hypotheses comprises displaying each word hypothesis at a larger size for larger importance values.
5. The method of claim 1 , wherein analyzing audio data comprises associating sets of related and competing hypotheses.
6. The method of claim 5 , wherein displaying the plurality of word hypotheses comprises visually emphasizing other hypotheses in a set of related hypotheses after a user selects a first hypothesis from the set of related hypotheses.
7. The method of claim 5 , wherein displaying the plurality of word hypotheses comprises visually deemphasizing other hypotheses in a set of competing hypotheses after a user selects a first hypothesis from the set of competing hypotheses.
8. A non-transitory computer readable storage medium comprising a computer readable program for information retrieval, wherein the computer readable program when executed on a computer causes the computer to perform the steps of:
analyzing audio data using a processor to produce a plurality of word hypotheses;
displaying the plurality of word hypotheses in motion at different respective speeds at once across a graphical display, the speeds being based on a confidence value associated with each respective word hypothesis; and
retrieve in accordance with one or more selected terms from the displayed word hypotheses.
9. A system for information retrieval, comprising:
a speech recognition module comprising a processor configured to analyze audio data to produce a plurality of word hypotheses;
a graphical user interface configured to display the plurality of word hypotheses in motion at different respective speeds at once across a graphical display, the speeds being based on a confidence value associated with each respective word hypothesis; and
an information retrieval module configured to retrieve information in accordance with one or more selected terms from the displayed word hypotheses.
10. The system of claim 9 , wherein each of the plurality of word hypotheses has an associated importance value.
11. The system of claim 10 , wherein only word hypotheses having an importance above an importance threshold are displayed.
12. The system of claim 10 , wherein the graphical user interface is further configured to display each word hypothesis at a larger size for larger importance values.
13. The system of claim 9 , wherein the speech recognition module is further configured to associate sets of related and competing hypotheses.
14. The system of claim 13 , wherein the graphical user interface is further configured to visually emphasize other hypotheses in a set of related hypotheses after a user selects a first hypothesis from the set of related hypotheses.
15. The system of claim 13 , wherein the graphical user interface is further configured to visually deemphasize other hypotheses in a set of competing hypotheses after a user selects a first hypothesis from the set of competing hypotheses.