Speech recognition system and method based on word-level candidate generation

A speech recognition system, comprising:

• a speech recognition server, the speech recognition server including a processor configured to,
• receive, via a sound signal receiver, a sound signal from a terminal, the sound signal being input through an input device of the terminal,
• recognize at least one recognized word during a recognition time period, the recognition time period being determined based on a time when recognition of the at least one recognized word is initiated, a time when recognition of the at least one recognized word is terminated, and an allowable time value added to the time when the recognition of the at least one recognized word is terminated,
• determine a word sequence corresponding to the sound signal and at least one candidate word for at least one word in the word sequence, the candidate word being selected from among the at least one recognized word by,
• selecting, from among the at least one recognized word, words that have a same starting time, and
• combining at least two of the recognized words to form the at least one candidate word, if some of the at least one candidate words have the same starting time and a shorter end time,
• generate a speech recognition result, the speech recognition result including the word sequence and the at least one candidate word, and
• provide the speech recognition result to the terminal, wherein
• the word sequence is displayed on the terminal such that the at least one word is visually distinguishable from other words of the word sequence, and
• in response to selection of the at least one word, the word sequence is displayed by sequentially replacing the at least one word with the at least one candidate word such that the different ones of the at least one candidate word may be substituted into the word sequence each time the at least one word is selected without displaying a list of the at least one candidate word.

A speech recognition method, comprising:

• verifying, by a processor, a word sequence and at least one candidate word for at least one word in the word sequence in response to receiving a speech recognition result from a speech recognition server, the speech recognition result including the word sequence and the at least one candidate word;
• displaying, by a display device connected to the processor, the word sequence such that the at least one word is visually distinguishable from other words of the word sequence,
• sequentially replacing the at least one word with the at least one candidate word in response to the at least one word being selected such that the different ones of the at least one candidate word may be substituted into the word sequence each time the at least one word is selected without displaying a list of the at least one candidate word, wherein,
• the at least one candidate word is selected among at least one recognized word by selecting, from among the at least one recognized word, words that have a same starting time,
• if some of the at least one candidate words have the same starting time and a shorter end time, then the speech recognition server is configured to combine at least two of the recognized words to form the at least one candidate word,
• the at least one recognized word is recognized by the speech recognition server according to the sound signal during a recognition time period, and
• the recognition time period being determined based on a time when recognition of the at least one recognized word is initiated, a time when recognition of the at least one recognized word is terminated, and an allowable time value added to the time when the recognition of the at least one recognized word is terminated.

A speech recognition method, comprising:

• receiving a sound signal from a terminal, the sound signal being input through an input device of the terminal,
• recognizing, by a processor, at least one recognized word during a recognition time period, the recognition time period being determined based on a time when recognition of the at least one recognized word is initiated, a time when recognition of the at least one recognized word is terminated, and a allowable time value added to the time when the recognition of the at least one recognized word is terminated,
• determining, by the processor, a word sequence corresponding to the sound signal and at least one candidate word for at least one word in the word sequence, the candidate word being selected from among the at least one recognized word by,
• selecting, from among the at least one recognized word, words that have a same starting time, and
• combining at least two of the recognized words to form the at least one candidate word, if some of the at least one candidate words have the same starting time and a shorter end time;
• generating, by the processor, the word sequence and the at least one candidate word as a speech recognition result;
• and
• providing the speech recognition result to the terminal, wherein
• the word sequence is displayed on the terminal such that the at least one word is visually distinguishable from other words of the word sequence, and
• in response to selection of the at least one word, the word sequence is displayed by sequentially replacing the at least one word with the at least one candidate word such that the different ones of the at least one candidate word may be substituted into the word sequence each time the at least one word is selected without displaying a list of the at least one candidate word.