é³å£°èªèï¼ããããã«ããããè±: speech recognitionï¼ã¯å£°ããã¤æ å ±ãã³ã³ãã¥ã¼ã¿ã«èªèãããã¿ã¹ã¯ã®ç·ç§°ã§ãã[1]ãããã®ï¼å¤©ç¶ï¼é³å£°èªèã¨å¯¾æ¯ãã¦èªåé³å£°èªèï¼è±: Automatic Speech Recognition; ASRï¼ã¨ãå¼ã°ãã[2]ã ä¾ã¨ãã¦æåèµ·ããã話è èªèãæããããã é³å£°èªèã¯ãé³å£°ã«å«ã¾ããæ å ±ãèªèããã¿ã¹ã¯ãã®ç·ç§°ã§ãããå ·ä½çã«è§£ãããåé¡ã®ä¾ã¨ãã¦ä»¥ä¸ãæããããï¼ Speech-to-Text (STT): å«ã¾ããè¨èªæ å ±ãæåã«å¤æããã¿ã¹ã¯ãããããæåèµ·ãã ãã¼ã¯ã¼ãèªèï¼è±èªçï¼(KWS): äºåã«è¨å®ããããã¼ã¯ã¼ãã®åºç¾ãèªèããã¿ã¹ã¯ãä¾ã¨ãã¦ããã¤ãSiriã é³å£°èªèããµãã¿ã¹ã¯ã¨ãã¦å«ãã¿ã¹ã¯ã«ã¯ä»¥ä¸ãæããããï¼ é³å£°æä½: é³å£°ã«ããã¢ããªã®æä½ãSST/KWSã§é³å£°æ å ±ãåãåºãããããã³ã³ã
{{#tags}}- {{label}}
{{/tags}}