Method of segmentation of chains of words in the training phase of a connected word recognizer

Method of segmentation of chains of words in the training phase of a connected word recognizer. Of the type that includes the steps of automatic segmentation 2 of the current chain of words to be segmented 1, comparison 3 of the number and identification of the words into which the current word chai...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: ESCRIHUELA LANGA FRANCISCO JOSE, RUIZ MEZCUA BELEN, GERBOLES ESPINA MARIA EUGENIA
Format: Patent
Sprache:eng ; spa
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Method of segmentation of chains of words in the training phase of a connected word recognizer. Of the type that includes the steps of automatic segmentation 2 of the current chain of words to be segmented 1, comparison 3 of the number and identification of the words into which the current word chain being segmented, has been segmented 1 with the values of the number of words and their identification corresponding to the original word chain 4, and decision 5 on the incorporation 8 of the segmentation obtained in Hidden Markov Models when comparison 3 shows that the number of words and their identification are the same. The method is characterized by the fact that when the previous comparison shows that the aforementioned values are not the same, a proportional segmentation 6 of the word chain being segmented 1 is carried out, consisting of dividing this chain into as many segments as there are words in the original word chain 4 and assigning to each segment a duration proportional to the duration of the corresponding word when it is pronounced in isolation 7. METODO DE SEGMENTACION DE CADENAS DE PALABRAS EN LA FASE DE ENTRENAMIENTO DE UN RECONOCEDOR DE PALABRAS CONECTADAS. DEL TIPO DE LOS QUE COMPRENDEN LAS ETAPAS DE SEGMENTACION AUTOMATICA (2) DE LA CADENA DE PALABRAS EN CURSO A SEGMENTAR (1), COMPARACION (3) DEL NUMERO E IDENTIFICACION DE LAS PALABRAS EN QUE SE HA SEGMENTADO LA CADENA DE PALABRAS EN CURSO A SEGMENTAR (1) CON LOS VALORES DEL NUMERO DE PALABRAS E IDENTIFICACION DE LAS MISMAS CORRESPONDIENTES A LA CADENA ORIGINAL DE PALABRAS (4), Y DECISION (5) SOBRE LA INCORPORACION (8) DE LA SEGMENTACION OBTENIDA A LOS MODELOS OCULTOS DE MARKOV CUANDO LA COMPARACION (3) MUESTRA QUE EL NUMERO DE PALABRAS Y LA IDENTIFICACION DE LAS MISMAS SON IGUALES. EL METODO SE CARACTERIZA PORQUE CUANDO LA COMPARACION ANTERIOR MUESTRA QUE LOS VALORES MENCIONADOS NO SON IGUALES, SE REALIZA UNA SEGMENTACION PROPORCIONAL (6) DE LA CADENA DE PALABRAS EN CURSO A SEGMENTAR (1), CONSISTENTE EN DIVIDIR DICHA CADENA EN TANTOS SEGMENTOS COMO PALABRAS CONTENGA LA CADENAORIGINAL DE PALABRAS (4) Y ASIGNAR A CADA SEGMENTO UNA DURACION PROPORCIONAL A LA DURACION DE LA PALABRA CORRESPONDIENTE CUANDO SE PRONUNCIA AISLADAMENTE (7).