TY - GEN
T1 - Efficient interactive weight tuning for tts synthesis
T2 - 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006
AU - Alías, Francesc
AU - Llorà, Xavier
AU - Formiga, Lluís
AU - Sastry, Kumara
AU - Goldberg, David E.
PY - 2006
Y1 - 2006
N2 - The quality of corpus-based text-to-speech systems depends on the accuracy of the unit selection process, which in turn relies on the cost function definition. This function should map the user perceptual preference when selecting synthesis units, which is a very difficult task. This paper continues our previous work on fusing the human judgements with the cost function by means of interactive weight tuning. The application of active interactive genetics algorithms mitigates user fatigue by improving user consistency. As a result, the obtained weights generate more natural synthetic speech when compared to previous objective and subjective proposals.
AB - The quality of corpus-based text-to-speech systems depends on the accuracy of the unit selection process, which in turn relies on the cost function definition. This function should map the user perceptual preference when selecting synthesis units, which is a very difficult task. This paper continues our previous work on fusing the human judgements with the cost function by means of interactive weight tuning. The application of active interactive genetics algorithms mitigates user fatigue by improving user consistency. As a result, the obtained weights generate more natural synthetic speech when compared to previous objective and subjective proposals.
UR - http://www.scopus.com/inward/record.url?scp=33947634558&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:33947634558
SN - 142440469X
SN - 9781424404698
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - I865-I868
BT - 2006 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings
Y2 - 14 May 2006 through 19 May 2006
ER -