DocumentCode
2214609
Title
Integration and evaluation of an HMM-based Text-To-Speech System to five
Author
Maciel, Alexandre ; Filho, Edson Carvalho
Author_Institution
Inf. Center, Fed. Univ. of Pernambuco, Pernambuco, Brazil
fYear
2012
fDate
11-13 April 2012
Firstpage
633
Lastpage
636
Abstract
This work presents an integration process and a quality evaluation of HMM-based Text-To-Speech System to FIVE (Framework for an Integrated Voice Environment). Speech synthesis approaches were studied, commercial synthesizers were tested and a standard of extensibility were developed for the FIVE. The integration process was successful and the quality evaluation was conducted based on three approaches: on MBROLA diphones, on HMM-based Text-To-Speech approach, and on Loquendo´s products. The acquired results were considered relevant for HMM voices analysis and for FIVE quality increment.
Keywords
hidden Markov models; speech synthesis; FIVE; HMM voices analysis; HMM-based text-to-speech system evalation; HMM-based text-to-speech system integration; Loquendo´s products; MBROLA diphones; commercial synthesizers; extensibility standard; framework for an integrated voice environment; hidden Markov models; integration process; quality evaluation; speech synthesis; Engines; Hidden Markov models; High temperature superconductors; Speech; Speech synthesis; Synthesizers; Training; Evaluation; Framework; Text-To-Speech;
fLanguage
English
Publisher
ieee
Conference_Titel
Systems, Signals and Image Processing (IWSSIP), 2012 19th International Conference on
Conference_Location
Vienna
ISSN
2157-8672
Print_ISBN
978-1-4577-2191-5
Type
conf
Filename
6208323
Link To Document