Method and system for Chinese large-vocabulary personal name automated speech recognition input

A method and system for Chinese large-vocabulary personal name automated speech recognition input, the method including the steps of: (a) having a user input a first speech sound to describe a personal name to be recognized, the personal name including plural characters; (b) using a Name Net Recogni...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: HUANG, LIANG-SHENG, SHEN, JIA-LIN, WANG, JUIANG, TSAI, JIM-HO
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:A method and system for Chinese large-vocabulary personal name automated speech recognition input, the method including the steps of: (a) having a user input a first speech sound to describe a personal name to be recognized, the personal name including plural characters; (b) using a Name Net Recognizer to recognize the first speech sound so as to obtain a recognition result for a personal name; (c) delivering the recognition result for a personal name to a Character Confirmation Unit; (d) using the Character Confirmation Unit to confirm each of the characters of the recognition result for a personal name; (e) if all characters are confirmed to be correct in step (d), the recognition result for a personal name is output after confirmation; (f) if one of the characters as recognized is confirmed to be incorrect in step (d), the user inputting a second speech sound in a type that could describe the incorrect character as recognized; (g) using a Description Recognition Unit corresponding to the type to recognize