Voice input system and voice input method

• US 8,463,609 B2
• Filed: 04/29/2009
• Issued: 06/11/2013
• Est. Priority Date: 04/30/2008
• Status: Active Grant

• Current Assignee: Delta Electronics Incorporated
• Original Assignee: Delta Electronics Incorporated
• Inventors: Huang, Liang-Sheng, Huang, Chao-Jen, Shen, Jia-Lin, Yeh, Keng-Hung
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US12/431,821
• Publication Number:
  US 20090276219A1
• Time in Patent Office: 1,504 Days
• Field of Search: None
• US Class Current: 704/251
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models

Abstract

In the present invention, a voice input system and a voice input method are provided. The voice input method includes the steps of: (A) initiating a speech recognition process by a first input associated with a first parameter of a first speech recognition subject; (B) providing a voice and a searching space constructed by a speech recognition model associated with the first speech recognition subject; (C) obtaining a sub-searching space from the searching space based on the first parameter; (D) searching at least one candidate item associated with the voice from the sub-searching space; and (E) showing the at least one candidate item.

Claims

1. A voice input system, comprising:

an input module generating a first input;

a voice module providing a voice;

a speech recognition model constructing a searching space;

a table having a plurality of inputs and a plurality of parameters corresponding to a first speech recognition subject associated with the speech recognition model, wherein the first speech recognition subject relates to one selected from a group consisting of a specific number of words, a specific number of syllables, a phrase having a specific number of terms and a combination thereof;

a processing module extracting from the table a first one of the plurality of parameters corresponding to the first input, obtaining a sub-searching space from the searching space based on the first parameter, and recognizing at least one candidate item corresponding to the voice from the sub-searching space through a speech recognition; and

an output module showing the at least one candidate item.

2. The voice input system according to claim 1, wherein the processing module receives a second input so as to further sift the at least one candidate item.

3. The voice input system according to claim 2, wherein the second input is associated with a second one of the plurality of parameters.

4. The voice input system according to claim 2, wherein the table further has a plurality of parameters corresponding to a second speech recognition subject having a corresponding parameter associated with the second input.

5. The voice input system according to claim 1, wherein the speech recognition model includes a model selected from a group consisting of an acoustic model, a lexical model, a language model and a combination thereof.

6. The voice input system according to claim 1, wherein the plurality of parameters are a plurality of numeral codes.

7. The voice input system according to claim 1, wherein the first speech recognition subject further relates to an input method being one selected from a group consisting of a phonetic notation input method, a phonetic transcription input method, a radical input method, a stroke input method and a tone input method, and the plurality of parameters are a plurality of corresponding codes of the input method.

8. The voice input system according to claim 1, wherein the first speech recognition subject further relates to one of a male voice and a female voice.

9. The voice input system according to claim 1, wherein the first speech recognition subject further relates to one of one of a rough acoustic model and an accurate acoustic model, and one of an anti-noise acoustic model and a non-anti-noise acoustic model.

10. The voice input system according to claim 1, wherein the first speech recognition subject further relates to a specific sentence pattern under a specific domain.

11. The voice input system according to claim 1, wherein the processing module further initiates a speech recognition function to perform the speech recognition according to the first input.

12. The voice input system according to claim 1, wherein the input module is one selected from a group consisting of a key, a keyboard, a virtual keyboard and a touch panel.

13. The voice input system according to claim 1, wherein the voice module is one selected from a group consisting of a voice input device, a voice storage device and a combination thereof.

14. The voice input system according to claim 1, wherein the at least one candidate item is at least one selected from a group consisting of a word, a phrase, a sentence and a combination thereof.

15. A voice input method, comprising steps of:

(A) initiating a speech recognition process by inputting a first input associated with a first parameter of a first speech recognition subject via an inputting device, wherein the first speech recognition subject relates to one selected from a group consisting of a specific number of words, a specific number of syllables, a phrase having a specific number of terms and a combination thereof;

(B) providing a voice and a searching space constructed by a speech recognition model associated with the first speech recognition subject;

(C) obtaining a sub-searching space from the searching space based on the first parameter;

(D) searching at least one candidate item associated with the voice from the sub-searching space; and

(E) showing the at least one candidate item.

16. The voice input method according to claim 15 further comprising a step of (F) sifting the at least one candidate item by a second input.

17. The voice input method according to claim 16, wherein the second input is associated with a corresponding parameter of the first speech recognition subject.

18. The voice input method according to claim 16, wherein the second input is associated with a second parameter of a second speech recognition subject.

19. A voice input method for a handset device, comprising steps of:

(A) providing a searching space based on a speech recognition model;

(B) receiving a first input parameter corresponding to a first speech recognition subject associated with the speech recognition model and obtaining a sub-searching space from the searching space based on the first input parameter, wherein the first speech recognition subject relates to one selected from a group consisting of a specific number of words, a specific number of syllables, a phrase having a specific number of terms and a combination thereof;

(C) receiving a voice and recognizing at least one candidate item associated with the voice from the sub-searching space through a speech recognition; and

(D) showing the at least one candidate item on the handset device.

20. The voice input method according to claim 19, wherein the speech recognition model comprises a plurality of speech recognition subjects, and the first input parameter is corresponding to the first speech recognition subject.

21. The voice input method according to claim 20 further comprising a step of (E) receiving a second input parameter so as to further sift the at least one candidate item accordingly.

22. A voice input method, having a searching space constructed by a speech recognition model, comprising steps of:

(A) receiving an input parameter corresponding to a first speech recognition subject associated with the speech recognition model and obtaining a sub-searching space from the searching space accordingly, wherein the first speech recognition subject relates to one selected from a group consisting of a specific number of words, a specific number of syllables, a phrase having a specific number of terms and a combination thereof;

(B) receiving a voice and recognizing at least one candidate item associated with the voice from the sub-searching space by a speech recognition; and

(C) showing the at least one candidate item by an outputting device.