Transliteration work support device, transliteration work support method, and computer program product

• US 9,928,828 B2
• Filed: 04/05/2016
• Issued: 03/27/2018
• Est. Priority Date: 10/10/2013
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Fume, Kosei, Kuroda, Yuka, Mizuoka, Yoshiaki, Morita, Masahiro
• Primary Examiner(s): Han, Qi
• Application Number:
  US15/090,776
• Publication Number:
  US 20160217782A1
• Time in Patent Office: 721 Days
• Field of Search: 704 9, 704 2, 704 3, 704 5, 704 7, 704258, 704266, 704277, 704278
• CPC Class Codes: G06F 40/129 Handling non-Latin characters, e.g. kana-to-kanji conversion
• CPC Class Codes: G06F 40/151 Transformation
• CPC Class Codes: G06F 40/232 Orthographic correction, e.g. spell checking or vowelisation
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G10L 13/027 Concept to speech synthesisers; Generation of natural phrases from machine-based concepts generation of parameters for speech synthesis out of text G10L13/08
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

Abstract

According to an embodiment, a transliteration work support device includes an analysis unit, a storage unit, an estimation unit, a construction unit, a correction unit, and an update unit. The analysis unit performs language analysis on document data and creates transliteration auxiliary information representing a way of transliteration of a word or a phrase in the document data. The storage unit stores a correction history representing a way of transliteration corrected in the past of the word or the phrase. The estimation unit estimates a correction place and a correction candidate of the document data or the transliteration auxiliary information from the history. The construction unit constructs work list information including work items corresponding to types of corrections according to the correction candidate and progress information. The correction unit corrects the document data or the transliteration auxiliary information. The update unit updates the history and the progress information according to the correction.

Claims

1. A transliteration work support device, comprising:

an analysis unit configured to perform language analysis on document data and create transliteration auxiliary information that represents a way of transliteration of a word or a;

phrase included in the document data in which voices of voice data are expressed for voice synthesis;

a storage unit configured to store therein a correction history that represents a way of transliteration, which has been corrected in the past, of the word or the phrase;

an estimation unit configured to estimate, from the correction history, (1) at least one of a correction place of the document data or a correction place of the transliteration auxiliary information, and (2) at least one of a correction candidate of the document data or a correction candidate of the transliteration auxiliary information;

a construction unit configured to construct work list information that includes work items corresponding to respective types of corrections according to the correction candidate of the document data or the correction candidate of the transliteration auxiliary information and includes progress information about the work items;

a correction unit configured to receive a correction instruction and correct the correction place of the document data or the correction place of the transliteration auxiliary information; and

an update unit configured to update the correction history and the progress information in the work list information in accordance with the correction performed by the correction unit,

wherein the construction unit and the update unit control the work list information to improve a quality of a voice content that is created when voice synthesis is performed using the document data and the transliteration auxiliary information.

2. The device according to claim 1, further comprising:

a display unit configured to display the work list information to an operator; and

a reception unit configured to receive the correction instruction by the operator.

3. The device according to claim 1, wherein

the work list information further includes quality information about the voice content that is to be created when the voice synthesis is performed using the document data and the transliteration auxiliary information, and

the update unit further updates the quality information in the work list information in accordance with the updated result of the document data or the transliteration auxiliary information.

4. The device according to claim 1, wherein the transliteration auxiliary information includes statistical information that represents an appearance frequency of the word or the phrase and logical element information that represents a layout of the word or the phrase included in the document data.

5. The device according to claim 4, wherein the statistical information includes a unique expression that expresses a land name, a person'"'"'s name, or a quantity expression and the statistical information about an unknown word or an unknown phrase that is not defined in dictionaries.

6. The device according to claim 4, wherein the construction unit calculates scores of the work items from the statistical information about the word or the phrase corresponding to the correction place of the document data or the correction place of the transliteration auxiliary information and from the type of correction according to the correction candidate of the document data or the correction candidate of the transliteration auxiliary information, and changes a display method of the work item based on the scores.

7. The device according to claim 6, wherein the construction unit highlights displays of a certain number of work items in a descending order of the scores.

8. The device according to claim 4, wherein the estimation unit estimates, further based on the logical element information, the correction place of the document data or the correction place of the transliteration auxiliary information, and the correction candidate of the document data or the correction candidate of the transliteration auxiliary information.

9. A transliteration work support method performed by a transliteration work support device, the method comprising:

performing language analysis on document data and creating transliteration auxiliary information that represents a way of transliteration of a word or a phrase included in the document data in which voices of voice data are expressed for voice synthesis;

referring to a correction history that represents a way of transliteration, which has been corrected in the past, of the word or the phrase;

estimating, from the correction history, (1) at least one of a correction place of the document data or a correction place of the transliteration auxiliary information, and (2) at least one of a correction candidate of the document data or a correction candidate of the transliteration auxiliary information;

constructing work list information that includes work items corresponding to respective types of corrections according to the correction candidate of the document data or the correction candidate of the transliteration auxiliary information and includes progress information about the work items;

receiving a correction instruction and correcting the correction place of the document data or the correction place of the transliteration auxiliary information;

updating the correction history and the progress information in the work list information in accordance with the correction at the correcting; and

controlling the work list information to improve a quality of a voice content that is created when voice synthesis is performed using the document data and the transliteration auxiliary information.

10. A computer program product comprising a non-transitory computer-readable medium including programmed instructions, the instructions causing a computer that includes a storage unit storing therein a correction history representing a way of transliteration, which has been corrected in the past, of a word or a phrase, to function as:

an analysis unit configured to perform language analysis on document data and create transliteration auxiliary information that represents a way of transliteration of a word or a phrase included in the document data in which voices of voice data are expressed for voice synthesis;

an estimation unit configured to estimate, from the correction history, (1) at least one of a correction place of the document data or a correction place of the transliteration auxiliary information, and (2) at least one of a correction candidate of the document data or a correction candidate of the transliteration auxiliary information;

a construction unit configured to construct work list information that includes work items corresponding to respective types of corrections according to the correction candidate of the document data or the correction candidate of the transliteration auxiliary information and includes progress information about the work items;

a correction unit configured to receive a correction instruction and correct the correction place of the document data or the correction place of the transliteration auxiliary information; and

an update unit configured to update the correction history and the progress information in the work list information in accordance with the correction performed by the correction unit,

wherein the construction unit and the update unit control the work list information to improve a quality of a voice content that is created when voice synthesis is performed using the document data and the transliteration auxiliary information.