Character recognizing and translating system and voice recognizing and translating system

• US 6,148,105 A
• Filed: 04/22/1999
• Issued: 11/14/2000
• Est. Priority Date: 11/15/1995
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Wakisaka, Shinji, Sato, Hiroko
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): Do, Anh Hong
• Application Number:
  US09/296,278
• Time in Patent Office: 572 Days
• Field of Search: 382/190, 382/203, 382/229, 704/254, 704/277, 704/248, 704/243, 704/253
• US Class Current: 382/190
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06F 40/53 Processing of non-Latin text kana-to-kanji conversion G06F40/129; vowelisation G06F40/232
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1444 Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence

Abstract

A study system of a voice recognizing and translating system is provided with a sound data base for storing data from which noise is removed; a sound analysis unit for extracting the features of the voice corresponding to the voice data stored in the sound data base; and a model learning unit for creating an acoustic model on the basis of the analysis result of the sound analysis unit. A recognition system of the voice recognizing and translating system is provided with: an acoustic model storing unit for storing acoustic models; a second sound analysis unit for extracting the feature of the voice corresponding to the data concerned on the basis of the data obtained by removing the data representing noise from the voice data of a newly input voice, and a voice collating unit for collating the voice data obtained by the second sound analysis unit with the data of the acoustic models so as to recognize the voice.

Claims

1. A voice recognizing and translating system for recognizing a voice and translating the voice into words or sentences, comprising:

an acoustic model generation unit which generates an acoustic model, and a recognizing and translating unit which recognizes and translates a voice using the acoustic model, wherein,

a) said acoustic model generation unit comprises;

a first noise deletion unit which removes noise data corresponding to noise from voice data representing a voice for an acoustic model;

a first sound analysis unit which extracts a feature of the voice corresponding to the voice data from which the noise data is removed by the first noise deletion unit;

a model learning unit which creates an acoustic model from the feature of the voice extracted by said first sound analysis unit; and

an acoustic model storing unit which stores the acoustic model created by the model learning unit in connection with the noise removed from the voice data by the first noise deletion unit; and

b) said recognizing and translating unit comprises;

a second noise deletion unit which removes noise data corresponding to noise from voice data representing a voice to be translated;

a second sound analysis unit which extracts a feature of the voice corresponding to the voice data from which the noise data is removed by the second noise deletion unit;

a voice collating unit which selects one acoustic model from said acoustic model storing unit based on the noise data removed from the voice data by the second noise deletion unit, and collates the feature of the voice extracted by said second sound analysis unit with the selected acoustic model to recognize the voice; and

a translation unit which translates words or sentences which are composed of the voice recognized by said voice collating unit.

2. A voice recognizing and translating system according to claim 1, wherein a result obtained by said translation unit is at least output as characters by a display device, and an original language text composed of words or sentences before being translated, and a language text composed of words or sentences after being translated, are displayed on the screen of said display device.

3. The voice recognizing and translating system as claimed in claim 2, wherein the result contains at least one of a translation rate representing the precision of the translation, or the type of language.

4. A voice recognizing and translating system for recognizing a detected voice and translating the voice into words or sentences, comprising:

a voice memory which stores voice data representing the detected voice;

a noise deletion unit which removes data corresponding to noise from the voice data;

a sound data base which stores the data from which the noise is removed by said noise deletion unit;

a first sound analysis unit which extracts the feature of a voice corresponding to the voice data stored in said sound data base;

a model learning unit which creates an acoustic model from the analysis result of said first sound analysis unit;

an acoustic model storing unit which stores the acoustic model;

a second sound analysis unit which extracts the feature of the voice corresponding to data which are obtained by removing the data representing noise from the voice data of the voice;

a voice collating unit which collates the voice data obtained by said second sound analysis unit with the data of the acoustic models stored in said acoustic model storing unit to recognize the detected voice; and

a translation unit which translates words or sentences which are composed of the detected voice recognized by said voice collating unit;

wherein said voice recognizing and translating system further comprises a stationary-mount information equipment having an external storage device, and a portable information equipment which is detachably connected to said stationary-mount type information equipment, and wherein said sound data base, said first sound analysis unit and said model learning unit are provided to said stationary-mount-type information equipment, said external storage device containing said sound data base, and all remaining constituent elements being provided to said portable information equipment.

5. A voice recognizing and translating system for recognizing a detected voice and translating the voice into words or sentences, comprising:

a voice memory which stores voice data representing the detected voice;

a noise deletion unit which removes data corresponding to noise from the voice data;

a sound data base which stores the data from which the noise is removed by said noise deletion unit;

a first sound analysis unit which extracts the feature of a voice corresponding to the voice data stored in said sound data base;

a model learning unit which creates an acoustic model from the analysis result of said first sound analysis unit;

an acoustic model storing unit which stores the acoustic model;

a second sound analysis unit which extracts the feature of the voice corresponding to data which are obtained by removing the data representing noise from the voice data of the voice;

a voice collating unit which collates the voice data obtained by said second sound analysis unit with the data of the acoustic models stored in said acoustic model storing unit to recognize the detected voice; and

a translation unit which translates words or sentences which are composed of the detected voice recognized by said voice collating unit;

wherein said memory is adapted to store first voice data corresponding to a first voice in which a surrounding noise is superposed on a target voice to be recognized and translated, and second voice data corresponding to a second voice composed of the surrounding noise.

6. The voice recognizing and translating system as claimed in claim 5, wherein said noise deletion unit compares the first voice spectral data of the first voice data with the second voice spectral data of the second voice data to obtain spectral data corresponding to the data from which the noise is removed.

7. The voice recognizing and translating system as claimed in claim 6, wherein said sound data base stores the first spectral data and the second spectral data while associating the same with each other.

8. The voice recognizing and translating system as claimed in claim 7, wherein said acoustic collating unit collates the noise spectral data of a surrounding noise with the data of a noise acoustic model obtained on the basis of the second spectral data, these data being obtained beforehand at a place where a voice to be recognized and translated is input, thereby recognizing the type of noise, and collates the voice data of the voice to be recognized and translated with the data of the acoustic model of the first voice data which is determined according to the type of noise, thereby recognizing the voice.

9. A voice recognizing and translating system for recognizing a detected voice and translating the voice into words or sentences, comprising:

a voice memory which stores voice data representing the detected voice;

a noise deletion unit which removes data corresponding to noise from the voice data;

a sound data base which stores the data from which the noise is removed by said noise deletion unit;

a first sound analysis unit which extracts the feature of a voice corresponding to the voice data stored in said sound data base;

a model learning unit which creates an acoustic model from the analysis result of said first sound analysis unit;

an acoustic model storing unit which stores the acoustic model;

a second sound analysis unit which extracts the feature of the voice corresponding to data which are obtained by removing the data representing noise from the voice data of the voice;

a voice collating unit which collates the voice data obtained by said second sound analysis unit with the data of the acoustic models stored in said acoustic model storing unit to recognize the detected voice; and

a translation unit which translates words or sentences which are composed of the detected voice recognized by said voice collating unit;

wherein said voice recognizing and translating system further comprises a stationary-mount type information equipment having an external storage device, and a portable information equipment which is detachably connected to said stationary-mount type information equipment, and wherein at least the sound data base is provided to said external storage device of said stationary-mount-type information equipment while all remaining constituent elements are provided to said portable information equipment.

10. A voice recognizing and translating system for removing noise data corresponding to noise from voice data representing an input voice so as to extract a feature of the voice corresponding to the voice data from which the noise data is removed and a feature of the noise corresponding to the noise data, creating an acoustic model on the basis of the feature of the voice and the feature of the noise, recognizing a newly input voice to be translated on the basis of the acoustic model, and translating words or sentences constituting the recognized voice, including:

a second sound analysis unit which extracts a feature of the newly input voice to be translated and extracts a feature of noise of the newly input voice;

a voice collating unit which collates the feature of the newly input voice extracted by said second sound analysis unit with the acoustic model corresponding to the feature of the noise extracted by said second sound analysis unit to recognize the newly input voice, wherein a different acoustic model is created for a feature of the input voice indicative of a same input voice if the feature of the noise differs; and

a translation unit which translates words or sentences constituting the newly input voice recognized by said voice collating unit.