Method and System for Voice Recognition Employing Multiple Voice-Recognition Techniques

• US 20140122071A1
• Filed: 02/22/2013
• Published: 05/01/2014
• Est. Priority Date: 10/30/2012
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Inventors: Labowicz, Michael P., Ondo, Kazuhiro, Sierawski, Jeffrey A., Bekkerman, Boris
• Granted Patent: US 9,570,076 B2
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• 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 2015/088 Word spotting
• CPC Class Codes: H04M 1/724 User interfaces specially adapted for cordless or mobile telephones
• CPC Class Codes: H04M 1/72436 for text messaging, e.g. short messaging services [SMS] or e-mails
• CPC Class Codes: H04M 2250/74 with voice recognition means

Abstract

A method and system for voice recognition are disclosed. In one example embodiment, the method includes receiving voice input information by way of a receiver on a mobile device and performing, by way of at least one processing device on the mobile device, first and second processing operations respectively with respect to first and second voice input portions, respectively, which respectively correspond to and are based at least indirectly upon different respective portions of the voice input information. The first processing operation includes a speech-to-text operation and the second processing operation includes an alternate processing operation. Additionally, the method includes generating recognized voice information based at least indirectly upon results from the first and second processing operations, and performing at least one action based at least in part upon the recognized voice information, where the at least one action includes outputting at least one signal by an output device.

Claims

1. A method of voice recognition comprising:

receiving first voice input information by way of a receiver;

converting by way of at least one processor the first voice input information or additional voice input information based at least indirectly upon the first voice input information into a first processed voice input portion;

performing by way of the at least one processor a word parsing operation with respect to the first processed voice input portion to identify at least two parsed word portions;

performing by way of the at least one processor at least one word spotting operation with respect to a first of the parsed word portions to generate a first revised word portion;

performing by way of the at least one processor a first speech-to-text operation with respect to a second of the parsed word portions or to a corresponding portion of the first voice input information, the additional voice input information, or the first processed voice input portion corresponding to the second parsed word portion, so as to arrive at a second revised word portion;

combining the first and second revised word portions to arrive at recognized voice information; and

performing at least one action based at least in part upon the recognized voice information, wherein the at least one action includes outputting at least one signal by way of an output device.

2. The method of voice recognition of claim 1, wherein the converting includes performing either the first speech-to-text operation or a second speech-to-text operation with respect to the first voice input information or the additional voice input information, and wherein the receiver includes a microphone.

3. The method of voice recognition of claim 2, wherein each of the at least two parsed word portions identified by the performing of the word parsing operation respectively includes at least one word.

4. The method of voice recognition of claim 3, wherein the at least two parsed word portions include at least three parsed word portions including the first and second word portions.

5. The method of voice recognition of claim 1, further comprising determining whether at least one of the at least two parsed word portions includes one or more words that correspond to a supported vocabulary.

6. The method of voice recognition of claim 1, further comprising determining that at least one of the at least two parsed word portions is an appropriate word portion with respect to which the at least one word spotting operation can be performed.

7. The method of voice recognition of claim 6, wherein the performing of the at least one word spotting operation includes consulting at least one memory device in which is stored at least one word set.

8. The method of voice recognition of claim 1, wherein the at least one word spotting operation includes at least one phonetic word spotting operation.

9. The method of voice recognition of claim 6, wherein the performing of the at least one word spotting operation includes:

first performing a first word spotting operation with respect to the first parsed word portion;

determining that the first performing of the first word spotting operation has failed to generate an acceptable revised word portion; and

second performing a second word spotting operation with respect to the first parsed word portion to generate the first revised word portion.

10. The method of voice recognition of claim 9, wherein the first word spotting operation includes first consulting at least one memory device to determine if the first parsed word portion corresponds to one or more first words of a first word set stored in the at least one memory device, and wherein the second word spotting operation includes second consulting the at least one memory device to determine if the first parsed word portion corresponds to one or more second words of a second word set stored in the at least one memory device.

11. The method of voice recognition of claim 10, wherein either at least one of the first and second word sets is a phone book stored in the at least one memory device, or the first parsed word portion is a name of a person.

12. The method of voice recognition of claim 1, further comprising either removing the first parsed word portion from the at least two parsed word portions or removing a further voice input portion corresponding to the first parsed word portion from the first voice input information, the additional voice input information, or the first processed voice input portion, so as to arrive at the corresponding portion of the first voice input information, the additional voice input information, or the first processed voice input portion, prior to the performing of the first speech-to-text operation with respect to the second parsed word portion or the corresponding portion.

13. The method of voice recognition of claim 1, wherein the outputting of the at least one signal by way of the output device includes one or more of:

displaying on a screen at least a portion of the recognized voice information;

providing a notification that an output message has been generated and is ready to be sent;

or

outputting by way of a speaker a response to the first voice input information.

14. The method of voice recognition of claim 13, wherein the output device is included in a mobile device that includes the receiver.

15. A method of voice recognition comprising:

receiving first voice input information by way of a receiver on a mobile device;

performing, by way of at least one processing device on the mobile device, first and second processing operations respectively with respect to first and second voice input portions, respectively, which respectively correspond to and are based at least indirectly upon different respective portions of the first voice input information,

wherein the first processing operation includes a speech-to-text operation and the second processing operation includes an alternate processing operation other than the speech-to-text operation;

generating recognized voice information based at least indirectly upon results from each of the first and second processing operations; and

performing at least one action based at least in part upon the recognized voice information, wherein the at least one action includes outputting at least one signal by way of an output device on the mobile device.

16. The method of voice recognition of claim 15, wherein the alternate processing operation includes a word spotting operation by which the second voice input portion is compared against a word set stored in at least one memory device of the mobile device.

17. The method of voice recognition of claim 16, wherein the word set is a phone book, and wherein the at least one action includes automatically generating a short message service (SMS) text message based at least indirectly upon the recognized voice information and either automatically sending the SMS text message or providing a notification for receipt by a user of the mobile device that the SMS text message is ready to be sent.

18. A system for voice recognition comprising:

at least one receiver configured to receive a first voice input signal;

at least one memory device configured to store at least one predefined word set;

at least one output device configured to provide an output signal; and

at least one processing device coupled at least indirectly to each of the at least one receiver, the at least one memory device, and the at least one output device,

wherein the at least one processing device is configured to perform first and second operations respectively upon respective first and second portions of voice information based at least indirectly upon the first voice input signal, the first operation including performing of a speech-to-text conversion with respect to the first portion of the voice information, and the second operation including performing of a word spotting operation with respect to the second portion the voice information,

wherein the at least one processing device is configured to combine results of the first and second operations at least indirectly to generate recognized voice information, and

wherein the at least one processing device is configured to cause the at least one output device to provide the output signal based at least in part upon the recognized voice information.

19. The system of claim 18, wherein the at least one processing device is further configured to perform a word parsing operation in relation to first voice input information based at least in part upon the voice input signal, so as to generate one or both of the first and second portions of the voice information.

20. The system of claim 18, wherein the system is a mobile device and the at least one output device of the mobile device includes one or more of a video display, a touch screen display, a speaker, a mechanical output device, and a wireless transmitter.