Evaluación del modelado acústico y prosódico del sistema de conversión texto-voz Cotovía

Unit selection speech synthesis has become the most researched area in speech technology, as a result of its high-quality performance. There are many approaches about the key points in this technology, that is, the cost functions design, speech parameterisation and prosody estimation, but there is very little research about suitable methods for evaluating the improvements in synthetic speech. In this work the subjective and objective tests performed on our TTS system Cotov́ıa are presented.