Speech recognition apparatus and system
AssigneeSorizava Co., Ltd.
InventorMunhak AN
Life of this patent
Estimating the term…
Abstract
The present invention relates to a speech recognition apparatus and system for recognizing speech and converting the speech into text, and displaying an input state thereof in real-time for correction. In the speech recognition apparatus, speech input from a speech input unit is converted into text in units of words to display the converted text in real time on a first display window, and words displayed on the first display window are combined to generate a sentence such that the generated sentence is displayed on a second display window in real time. Therefore, a process through which what kind of sentence is formed by a combination of what kind of words may be intuitively confirmed so that text generated through speech recognition may be easily corrected.
Claims11
A speech recognition apparatus comprising:
- a speech input unit configured to receive speech;
- a key input unit configured to receive a signal for correcting a word or a sentence converted into text through the speech input from the speech input unit;
- a display unit including a first display window configured to display words obtained by converting the speech input from the speech input unit into text and a second display window configured to display a sentence obtained by combining the words displayed on the first display window;
- and
- a controller configured to convert the speech input from the speech input unit into text in units of words and allow the converted text to be displayed in real time on the first display window, and combine the words displayed on the first display window to generate a sentence and allow the generated sentence to be displayed in real time on the second display window,
- wherein:
- the controller allows a first cursor, which is displayed at an end of a word displayed in real time on the first display window and indicates an input state of the word, to be displayed on the first display window, allows a second cursor, which is displayed at an end of a combined sentence displayed in real time on the second display window and indicates an input state of the sentence, to be displayed on the second display window, and allows a third cursor for correction according to a signal of the key input unit to be moved and displayed on the first display window or the second display window;
- and
- the controller displays as many special characters as the number of characters of a word that cannot be recognized among the words displayed on the first display window or the second display window to identify the number of characters.
A speech recognition system comprising:
- a content providing server configured to provide a content including speech;
- and
- a speech recognition apparatus including a speech input unit configured to receive content from the content providing server and receive the speech included in the content, a key input unit configured to receive a signal for correcting a word or a sentence converted into text through the speech input from the speech input unit, a display unit including a first display window configured to display words obtained by converting the speech input from the speech input unit into text and a second display window configured to display a sentence obtained by combining the words displayed on the first display window, and a controller configured to convert the speech input from the speech input unit into text in units of words and allow the converted text to be displayed in real time on the first display window, and combine the words displayed on the first display window to generate a sentence and allow the generated sentence to be displayed in real time on the second display window,
- wherein:
- the controller allows a first cursor, which is displayed at an end of a word displayed in real time on the first display window and indicates an input state of the word, to be displayed on the first display window, allows a second cursor, which is displayed at an end of a combined sentence displayed in real time on the second display window and indicates an input state of the sentence, to be displayed on the second display window, and allows a third cursor for correction according to a signal of the key input unit to be moved and displayed on the first display window or the second display window;
- and
- the controller displays as many special characters as the number of characters of a word that cannot be recognized among the words displayed on the first display window or the second display window to identify the number of characters.
Dependencies are inferred from the claim text.