Allowing spelling of arbitrary words

• US 9,971,758 B1
• Filed: 01/06/2016
• Issued: 05/15/2018
• Est. Priority Date: 01/06/2016
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google LLC (Alphabet Inc.)
• Inventors: Cherepanov, Evgeny A., Skobeltsyn, Gleb, Foerster, Jakob Nicolaus, Aleksic, Petar, Michaely, Assaf Avner Hurwitz
• Primary Examiner(s): Patel, Shreyans
• Application Number:
  US14/989,621
• Time in Patent Office: 860 Days
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G10L 15/187 Phonemic context, e.g. pronunciation rules, phonotactical constraints or phoneme n-grams
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems
• CPC Class Codes: G10L 2015/086 Recognition of spelled words
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for natural language processing. One of the methods includes receiving a first voice input from a user device; generating a first recognition output; receiving a user selection of one or more terms in the first recognition output; receiving a second voice input spelling a correction of the user selection; determining a corrected recognition output for the selected portion; and providing a second recognition output that merges the first recognition output and the corrected recognition output.

Claims

1. A method comprising:

receiving a first voice input from a user device;

generating a first recognition output;

receiving a user selection of one or more terms in the first recognition output;

receiving a second voice input spelling a correction of the user selection, the second voice input including a series of letters;

determining a corrected recognition output for the selected portion, wherein determining a corrected recognition output includes;

initiating parallel voice recognition processes on the second voice input including a letter specific voice recognition process and a base voice recognition process,

scoring recognition outputs generated for each of the parallel voice recognition processes, wherein the letter specific voice recognition process is scored based on the recognition of the spelled correction and wherein the base voice recognition process is scored based on the recognition for the spelled correction and one or more other terms of the first recognition output, and

selecting the corrected recognition output based on the scores; and

providing a second recognition output that merges the first recognition output and the corrected recognition output.

2. The method of claim 1, further comprising:

obtaining search results responsive to the second recognition output; and

providing one or more of the search results to the user device.

3. The method of claim 1, wherein the letter specific voice recognition process is configured to score an arbitrary sequence of letters.

4. The method of claim 1, wherein the base voice recognition process is used to generate the first recognition output.

5. The method of claim 1, wherein merging the first recognition output and the corrected recognition output includes replacing the user selection of the first recognition output with the corrected recognition output.

6. A system comprising:

one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising;

receiving a first voice input from a user device;

generating a first recognition output;

receiving a user selection of one or more terms in the first recognition output;

receiving a second voice input spelling a correction of the user selection, the second voice input including a series of letters;

determining a corrected recognition output for the selected portion, wherein determining a corrected recognition output includes;

initiating parallel voice recognition processes on the second voice input including a letter specific voice recognition process and a base voice recognition process,

scoring recognition outputs generated for each of the parallel voice recognition processes, wherein the letter specific voice recognition process is scored based on the recognition of the spelled correction and wherein the base voice recognition process is scored based on the recognition for the spelled correction and one or more other terms of the first recognition output, and

selecting the corrected recognition output based on the scores; and

providing a second recognition output that merges the first recognition output and the corrected recognition output.

7. The system of claim 6, wherein the one or more computers are further operable to perform operations comprising:

obtaining search results responsive to the second recognition output; and

providing one or more of the search results to the user device.

8. The system of claim 6, wherein the letter specific voice recognition process is configured to score an arbitrary sequence of letters.

9. The system of claim 6, wherein the base voice recognition process is used to generate the first recognition output.

10. The system of claim 6, wherein merging the first recognition output and the corrected recognition output includes replacing the user selection of the first recognition output with the corrected recognition output.

11. One or more non-transitory computer-readable medium encoded with instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:

receiving a first voice input from a user device;

generating a first recognition output;

receiving a user selection of one or more terms in the first recognition output;

receiving a second voice input spelling a correction of the user selection, the second voice input including a series of letters;

determining a corrected recognition output for the selected portion, wherein determining a corrected recognition output includes;

initiating parallel voice recognition processes on the second voice input including a letter specific voice recognition process and a base voice recognition process,

scoring recognition outputs generated for each of the parallel voice recognition processes, wherein the letter specific voice recognition process is scored based on the recognition of the spelled correction and wherein the base recognition process is scored based on the recognition for the spelled correction and one or more other terms of the first recognition output, and

selecting the corrected recognition output based on the scores; and

providing a second recognition output that merges the first recognition output and the corrected recognition output.

12. The non-transitory computer-readable medium of claim 11, wherein the instruction further cause the one or more computers to perform operations comprising:

obtaining search results responsive to the second recognition output; and

providing one or more of the search results to the user device.

13. The non-transitory computer-readable medium of claim 11, wherein the letter specific voice recognition process is configured to score an arbitrary sequence of letters.

14. The non-transitory computer-readable medium of claim 11, wherein merging the first recognition output and the corrected recognition output includes replacing the user selection of the first recognition output with the corrected recognition output.