IP Library Granted Patent US 10,650,810
Granted Patent B2
US 10,650,810 · App. 16/342,718 · Granted May 12, 2020

Determining phonetic relationships

Inventors: Nikhil Chandru Rao (Santa Clara, CA); Saisuresh Krishnakumaran (Mountain View, CA)
Assignee: GOOGLE LLC
G10L15/187G06F40/194G06F40/289G10L13/00G10L15/22G10L15/26G10L15/30G10L25/51
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,650,810
App. No.
16/342,718
Granted
May 12, 2020
Kind
B2
Abstract

Systems and methods of determining phonetic relationships are provided. For instance data indicative of an input text phrase input by a user can be received. An audio output corresponding to a spoken rendering of the input text phrase can be determined. A text transcription of the audio output of the input text phrase can be determined. The text transcription can be a textual representation of the audio output. The text transcription can be compared against a plurality of test phrases to identify a match between the text transcription and at least one test phrase.

Claims (43)

1. A computer-implemented method of determining a phonetic relationship between two or more phrases, the method comprising:

receiving, by one or more computing devices, data indicative of an input text phrase by a user;

determining, by the one or more computing devices, an audio output corresponding to a synthesized spoken rendering of the input text phrase;

determining, by the one or more computing devices, a text transcription of the audio output of the input text phrase, the text transcription comprising a textual representation of the audio output;

comparing, by the one or more computing devices, the text transcription against a plurality of test phrases;

identifying, by the one or more computing devices and based at least in part on the comparing, a match between the text transcription and a test phrase of the plurality of test phrases; and

responsive to identifying the match, determining, by the one or more computing devices, a phonetic distance between the text transcription of the audio output and the input text phrase,

wherein determining the phonetic distance is based on a first phonetic transcription associated with the input text phrase and a second phonetic transcription associated with the text transcription.

2. The computer-implemented method of claim 1 , further comprising, responsive to identifying the match, providing, by the one or more computing devices, an indication that the input text phrase is phonetically similar to the test phrase.

3. The computer-implemented method of claim 1 , further comprising, when the phonetic distance between the text transcription and input text phrase is less than a threshold, providing, by the one or more computing devices, an indication that the input text phrase is phonetically similar to the test phrase.

4. The computer-implemented method of claim 1 , wherein determining the phonetic distance based on the first and second phonetic transcriptions comprises:

determining, by the one or more computing devices, the first and second phonetic transcriptions, the first and second phonetic transcriptions each comprising a plurality of phonetic speech units; and

determining, by the one or more computing devices, a number of phonetic speech units in the second phonetic transcription that are different from the first phonetic transcription.

5. The computer-implemented method of claim 1 , wherein determining, by the one or more computing devices, the text transcription of the audio output of the input text phrase comprises determining the text transcription based at least in part on one or more language models.

6. The computer-implemented method of claim 5 , wherein the one or more language models comprise a biased language model determined based at least in part on a general language model and the plurality of test phrases.

7. The computer-implemented method of claim 1 , wherein the audio output of the text input phrase comprises a speech waveform corresponding to the synthesized spoken rendering of the input text phrase.

8. The computer-implemented method of claim 1 , wherein determining the audio output corresponding to the synthesized spoken rendering of the input text phrase comprises:

using speech synthesis to convert the input text phrase to the audio output.

9. The computer-implemented method of claim 1 , wherein the user is a developer of a computing application, and wherein the computing application is associated with a computing platform.

10. The computer-implemented method of claim 9 , wherein the plurality of test phrases include at least a list of reserved voice commands for the computing platform.

11. A computing system, comprising:

one or more processors; and

one or more memory devices, the one or more memory devices storing computer-readable instructions that when executed by the one or more processors cause the one or more processors to perform operations, the operations comprising:

receiving data indicative of an input text phrase input by a user;

determining an audio output corresponding to a synthesized spoken rendering of the input text phrase;

determining a text transcription of the audio output of the input text phrase, the text transcription comprising a textual representation of the audio output;

comparing the text transcription against a plurality of test phrases;

identifying, based at least in part on the comparing, a match between the text transcription and a test phrase of the plurality of test phrases; and

responsive to identifying the match, determining a phonetic distance between the text transcription of the audio output and the input text phrase,

wherein determining the phonetic distance is based on a first phonetic transcription associated with the input text phrase and a second phonetic transcription associated with the text transcription.

12. The computing system of claim 11 , the operations further comprising, responsive to identifying the match, providing an indication that the input text phrase is phonetically similar to the test phrase.

13. The computing system of claim 11 , the operations further comprising, when the phonetic distance between the text transcription and the input text phrase is less than a threshold, providing an indication that the input text phrase is phonetically similar to the text phrase.

14. The computing system of claim 11 , wherein the operations, in determining a the phonetic distance based on the first and second phonetic transcriptions, further comprising:

determining the first and second phonetic transcriptions, the first and second phonetic transcriptions each comprising a plurality of phonetic speech units; and

determining a number of phonetic speech units in the second phonetic transcription that are different from the first phonetic transcription.

15. One or more tangible, non-transitory computer-readable media storing computer-readable instructions that when executed by one or more processors cause the one or more processors to perform operations, the operations comprising:

receiving data indicative of an input text phrase input by a user;

determining an audio output corresponding to a synthesized spoken rendering of the input text phrase;

determining a text transcription of the audio output of the input text phrase, the text transcription comprising a textual representation of the audio output;

comparing the text transcription against a plurality of test phrases;

identifying, based at least in part on the comparing, a match between the text transcription and a test phrase of the plurality of test phrases; and

responsive to identifying the match, determining a phonetic distance between the text transcription of the audio output and the input text phrase,

wherein determining the phonetic distance is based on a first phonetic transcription associated with the input text phrase and a second phonetic transcription associated with the text transcription.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 24, 2019
From: RAO, NIKHIL CHANDRU; KRISHNAKUMARAN, SAISURESH
To: GOOGLE INC.
Reel/Frame 048985/0304 →
CHANGE OF NAME Recorded Apr 24, 2019
From: GOOGLE INC.
To: GOOGLE LLC
Reel/Frame 048989/0705 →
Continuity (2)
Provisional Application 62410564 · Oct 20, 2016
Related Publication 20190295531A1 · Sep 26, 2019