Method and apparatus for text input utilizing speech recognition

• US 20020091520A1
• Filed: 11/20/2001
• Published: 07/11/2002
• Est. Priority Date: 11/22/2000
• Status: Abandoned Application

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Nishizaki, Makoto, Endo, Mitsuru, Saito, Natsuki
• Application Number:
  US09/989,561
• Publication Number:
  US 20020091520A1
• Time in Patent Office: Days
• US Class Current: 704/235
• CPC Class Codes: G10L 15/19 Grammatical context, e.g. disambiguation of the recognition hypotheses based on word sequence rules

Abstract

A sentence or text based input utterance is prepared into word-string candidates each based on one to several words, starting from the beginning of a sentence. The candidates are displayed. A user is allowed to successively select and fix a candidate, to prepare a candidate for the following utterance on the basis of the selected candidate. The present invention is a text input method and apparatus adapted to repeat these processes. This eliminates the necessity to keep a memory space for search taking account of a number of combinations of word-string candidates, thus greatly reducing the storage capacity. Reduction is also possible in speech-recognition process amount. These makes possible apparatus size reduction. Furthermore, the user who is allowed to input one or more continuous utterances is free from troublesomeness as encountered in word-based input.

Claims

1. A method for inputting a text comprising:

(a) a step of continuously inputting an utterance;

(b) a step of preparing word-string candidates based on one to several words, staring from a beginning of the inputted utterance;

(c) a step of displaying the candidates; and

(d) a step of selecting the displayed candidate by a user;

whereby, for a following utterance, said candidate preparing step (b), said displaying step (c) and said selecting step (d) are repeated in order on the basis of the selected candidate.

2. A method for inputting a text according to claim 1, wherein said candidate preparing step (b) determines a phrase-based candidate by an extension process to repeat word linking according to a word-based linkage probability.

3. A method for inputting a text according to claim 2, wherein said candidate preparing step (b) further having a process to update the candidate due to an acoustic score.

4. A method for inputting a text according to claim 3, wherein said extension process is ended by reaching of the number of phrase candidates subjected to said extension process a predetermined number as counted from a top rank in a language score.

5. An apparatus for inputting a text comprising:

an input section for inputting an utterance;

an utterance pre-processing section for extracting a feature amount of an utterance of from said input section;

a word candidate preparing section for preparing a following word candidate from a fixed word string by using a language model;

a word-string preparing section for preparing a word-string candidate based on one to several words from the extracted feature amount and the word candidate by using at least any one of a language model and an acoustic model;

a display section for displaying the word-string candidate;

an operating section for a user to select the word-string candidate being displayed; and

a candidate-preparation instructing section for instructing said word candidate preparing section to prepare a following word candidate from a word string selected by said operating section.

6. An apparatus for inputting a text according to claim 5, wherein said word-string preparing section prepares a phrase-based candidate by an extension process to repeat word linking according to word-based linkage probability.

7. An apparatus for inputting a text according to claim 6, wherein said word-string preparing section further having an updating process due to an acoustic score.

8. An apparatus for inputting a text according to claim 7, wherein said word-string preparing section ends the extension process by reaching of the number of phrase candidate subjected to the extension process a predetermined number as counted from a top rank in a language score.

9. An apparatus for inputting a text according to claim 5, wherein said apparatus is included in a cellular telephone.

10. An apparatus for inputting a text according to claim 6, wherein said apparatus is included in a cellular telephone.

11. An apparatus for inputting a text according to claim 7, wherein said apparatus is included in a cellular telephone.

12. An apparatus for inputting a text according to claim 8, wherein said apparatus is included in a cellular telephone.

14. A storage medium for providing a program according to claim 13, wherein said candidate preparing step (b) determines a phrase-based candidate by an extension process to repeat word linking according to a word-based linkage probability.

15. A storage medium for providing a program according to claim 14, wherein said candidate preparing step (b) further having a process to update the candidate due to an acoustic score.

16. A storage medium for providing a program according to claim 15, wherein said extension process is ended by reaching of the number of phrase candidates subjected to said extension process a predetermined number as counted from a top rank in a language score.

18. A computer program product according to claim 17, wherein said candidate preparing step (b) determines a phrase-based candidate by an extension process to repeat word linking according to a word-based linkage probability.

19. A computer program product according to claim 18, wherein said candidate preparing step (b) further having a process to update the candidate due to an acoustic score.

20. A computer program product according to claim 19, wherein said extension process is ended by reaching of the number of phrase candidates subjected to said extension process a predetermined number as counted from a top rank in a language score.

13. A storage medium for providing a program to repeat in order:

(a) a step of continuously inputting an utterance;

(b) a step of preparing word-string candidates based on one to several words, staring from a beginning of the inputted utterance;

(c) a step of displaying the candidates; and

(d) a step of selecting the displayed candidate by a user;

whereby, for a following utterance, said candidate preparing step (b), said displaying step (c) and said selecting step (d) are repeated in order on the basis of the selected candidate.

17. A computer program product to repeat in order:

(a) a step of continuously inputting an utterance;

(b) a step of preparing word-string candidates based on one to several words, staring from a beginning of the inputted utterance;

(c) a step of displaying the candidates; and

(d) a step of selecting the displayed candidate by a user;

whereby, for a following utterance, said candidate preparing step (b), said displaying step (c) and said selecting step (d) are repeated in order on the basis of the selected candidate.