Reminder setting method and apparatus

A reminder setting apparatus, comprising:

• at least a processor, a microphone, a display, and at least a non-transitory computer readable storage medium storing instructions of programs executed by the processor, wherein the programs comprise:
• a speech acquiring module, configured to acquire a speech signal using the microphone and extract, from the speech signal, a plurality of speech segments;
• a first recognizing module, configured to apply the plurality of speech segments to a predefined keyword search network including foreground models and background models, each foreground model configured to recognize a time-specific keyword from a language and each background model configured to a non-time-specific keyword from the language, and acquire time information in the speech signal acquired by the speech acquiring module when one or more of the plurality of speech segments are identified by the predefined keyword search network as containing a time-specific keyword associated with one of the foreground models, the time information including the time-specific keyword, and start and end times of the identified speech segments in the speech signal;
• a time determining module, configured to determine a reminder time for reminder setting according to the time information acquired by the first recognizing module;
• a second recognizing module, configured to acquire, by using continuous speech recognition, a text sequence corresponding to the speech signal acquired by the speech acquiring module;
• a content determining module, configured to determine reminder content according to the time information acquired by the first recognizing module and the text sequence acquired by the second recognizing module;
• and
• a reminder setting module, configured to set a reminder according to the reminder time determined by the time determining module and the reminder content determined by the content determining module and display, on the display, the reminder including the reminder time, the reminder content and an option to replay the speech signal.

A non-transitory computer readable storage medium storing one or more programs in connection with a computing device having at least a processor, a microphone, and a display, which, when executed by the processor, performs the steps of:

• acquiring a speech signal using the microphone;
• extracting, from the speech signal, a plurality of speech segments;
• applying the plurality of speech segments to a predefined keyword search network including foreground models and background models, each foreground model configured to recognize a time-specific keyword from a language and each background model configured to a non-time-specific keyword from the language;
• acquiring time information in the speech signal when one or more of the plurality of speech segments are identified by the predefined keyword search network as containing a time-specific keyword associated with one of the foreground models, the time information including the time-specific keyword, and start and end times of the identified speech segments in the speech signal, and determining a reminder time for reminder setting according to the time information;
• acquiring a text sequence corresponding to the speech signal by using continuous speech recognition, and determining reminder content for reminder setting according to the time information and the text sequence;
• setting a reminder according to the reminder time and the reminder content;
• and
• displaying, on the display, the reminder including the reminder time, the reminder content and an option to replay the speech signal.