Speech synthesis unit selection

Speech synthesis unit selection

A Text To Speech (TTS) system 116 receives text data 134, parses it into sequences of text units 118 and determines multiple paths of corresponding speech units (eg. phonemes) 200 by selecting from the speech corpus 124 a number of second speech units (130a, 130b) which may be concatenated with a fi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Ioannis Agiomyrgiannakis
Format: Patent
Sprache:eng
Schlagworte:
ACOUSTICS
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:A Text To Speech (TTS) system 116 receives text data 134, parses it into sequences of text units 118 and determines multiple paths of corresponding speech units (eg. phonemes) 200 by selecting from the speech corpus 124 a number of second speech units (130a, 130b) which may be concatenated with a first speech unit 128 based on a join cost (eg. how well its acoustic characteristics fit with those of the first unit) and a target cost (representing the units accuracy in representing the phonetic unit). One such path is then selected as sounding most natural for audible output at a speech synthesizer.