Neural network for word recognition system with automatic learning capability - has characteristic vectors identified from speech signal that are classified based upon defined control regions

The neural network is structured with 3 layers with the first providing processing of inputs to generate token vectors that are stored in the second level after classification. The binary output signals of the intermediate neurons are transferred to output neuron layer elements. The classification p...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	BOTHE, ACHIM., 4790 PADERBORN, DE, BARSCHDORFF, DIETER, DR.., 4790 PADERBORN, DE, GAERTNER, ULRICH., 4790 PADERBORN, DE, JAEGER, ANDREAS., 4790 PADERBORN, DE
Format:	Patent
Sprache:	eng ; ger
Schlagworte:	ACOUSTICS CALCULATING COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS COMPUTING COUNTING MUSICAL INSTRUMENTS PHYSICS SPEECH ANALYSIS OR SYNTHESIS SPEECH OR AUDIO CODING OR DECODING SPEECH OR VOICE PROCESSING SPEECH RECOGNITION
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	The neural network is structured with 3 layers with the first providing processing of inputs to generate token vectors that are stored in the second level after classification. The binary output signals of the intermediate neurons are transferred to output neuron layer elements. The classification process uses defined control regions that can be defined of various sizes. All learning vectors are tested for presence within the regions and sub classes can be defined. USE/ADVANTAGE - Reduces processing time of learning process. Can be used with speech systems.