Language data storage and reproduction apparatus

• US 6,026,407 A
• Filed: 11/27/1996
• Issued: 02/15/2000
• Est. Priority Date: 11/28/1995
• Status: Expired due to Fees

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Hanada, Takako, Muraki, Kazunori
• Primary Examiner(s): AMSBURY, WAYNE P
• Application Number:
  US08/757,869
• Time in Patent Office: 1,175 Days
• Field of Search: 707/102, 707/2, 704/270
• US Class Current: 1/1
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G11B 2220/90 Tape-like record carriers
• CPC Class Codes: G11B 27/107 of operating tapes
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

The invention provides a language data storage and reproduction apparatus wherein manual application of a tag is unnecessary and, even if key inputting for retrieval is not accurate, object language data can be reproduced. A tag application section automatically applies to each language data inputted from a language data inputting section a tag representative of a distribution position of the language data in the entire language data, and a data recognition section performs data recognition processing of the individual language data to obtain recognition data. A storage section stores the recognition data together with the language data to which the tags are applied with reference to the tags. If a word or a phrase to be retrieved is inputted, a key retrieval section retrieves recognition data coincident with the word or the phrase from within the storage section. When the tag corresponding to the retrieved recognition data is outputted as a retrieval result, a data extraction section extracts the language data designated by the tag from the storage section. A language data reproduction section reproduces the language data extracted by the data extraction section.

Claims

1. A language data storage and reproduction apparatus, comprising:

a language data inputting section for inputting a series of language data as an electric signal string;

a tag application section for automatically applying to each of the language data inputted from said language data inputting section a tag representative of a distribution position of the language data in the entire series of language data;

a data recognition section for producing a plurality of recognition data sets, wherein each recognition data set corresponds to a portion of said electric signal string, and each recognition data set contains a plurality of candidates for recognition of said language data, wherein each of the candidates of one recognition data set corresponds to the same portion of said electric signal string as the other candidates in said one recognition data set, and wherein each of the candidates of said one recognition data set is different from any other candidate in said one recognition data set;

a storage section for storing the recognition data sets together with the language data to which the tags are applied with reference to the tags;

a retrieval key inputting section for inputting a word or a phrase by means of keys;

a key retrieval section for retrieving that one of the recognition data sets having a candidate which is coincident with the word or phrase inputted from said retrieval key inputting section from within said storage section and outputting the tag corresponding to the retrieved recognition data set as a retrieval result;

a data extraction section for referring to the tag of the retrieval result of said key retrieval section to extract the language data designated by the tag from said storage section; and

a language data reproduction section for reproducing the language data extracted by said data extraction section.

2. A language data storage and reproduction apparatus as claimed in claim 1, wherein said key retrieval section outputs, as the retrieval result, a tag representative of a start point and another tag representative of an end point of the recognition data set, and said data extraction section extracts the language data delineated by the tags from said storage section.

3. A language data storage and reproduction apparatus as claimed in claim 1, wherein the language data are speech data, and said data recognition section produces a plurality of candidates of speech recognition data by speech recognition whereas said storage section stores the plurality of candidates of speech recognition data together with the language data with the tag applied thereto with reference to the tag.

4. A language data storage and reproduction apparatus as claimed in claim 1, wherein the language data are image data, and said data recognition section produces a plurality of candidates of character recognition data by character recognition whereas said storage section stores the plurality of candidates of character recognition data together with the language data with the tag applied thereto with reference to the tag.

5. A language data storage and reproduction apparatus comprising:

a language data inputting section for inputting an electric signal string representative of language data;

a data recognition section for producing a plurality of recognition data sets, wherein each recognition data set corresponds to a portion of said electric signal string, and each recognition data set contains a plurality of candidates for recognizing said language data, wherein each of the candidates of one recognition data set corresponds to the same portion of said electric signal string as the other candidates in said one recognition data set, and wherein each of the candidates of said one recognition data set is different from any other candidate in said one recognition data set;

a storage section for storing each recognition data set together with the corresponding portion of said electric signal string;

an inputting section for inputting a desired word or phrase to be retrieved;

a retrieval section for comparing the desired word or phrase to said plurality of candidates in said recognition data sets, and for retrieving the recognition data set which contains the candidate most closely matched to the desired word or phrase;

a reproduction section for reproducing the segment of language data which corresponds to the recognition data set retrieved by the retrieval section.

6. A language data storage and reproduction apparatus as claimed in claim 5, wherein said electric signal string is a waveform representative of audio data, and said inputting section includes a microphone.

7. A language data storage and reproduction apparatus as claimed in claim 5, wherein said reproduction section includes a speaker.

8. A language data storage and reproduction apparatus as claimed in claim 5, further comprising a tag application section for automatically applying a tag to each portion of said electric signal string, wherein each tag is representative of the position of its corresponding electric signal string portion within the entire electric signal string.

9. A language data storage and reproduction apparatus as claimed in claim 8, wherein said storage section stores each recognition data set together with the corresponding electric signal string portion and associated tag.

10. A language data storage and reproduction apparatus as claimed in claim 8, wherein when said retrieval section retrieves the recognition data set which contains the candidate most closely matched to the desired word or phrase, said retrieval section also retrieves the tag associated with the corresponding electric signal string portion.

11. A language data storage and reproduction apparatus as claimed in claim 10, further comprising an extraction section which extracts from said storage section the electric signal string portion associated with the retrieved tag, for reproduction by said reproduction section.

12. A language data storage and reproduction apparatus as claimed in claim 10, wherein said retrieved tag is representative of the start point and the end point of the retrieved recognition data set as well as the start point and end point of the corresponding electric signal string portion.

13. A language data storage and reproduction apparatus as claimed in claim 8, wherein said electric signal string is representative of image data, and each of said tags applied by said tag application section is representative of the position of the corresponding portion of said electric signal string in a two dimensional coordinate system associated with the entire electric signal string.

14. A language data storage and reproduction apparatus as claimed in claim 5, wherein said electric signal string is representative of image data.

15. A language data storage and reproduction apparatus as claimed in claim 5, wherein said inputting section for inputting a desired word or phrase includes a keyboard.