Sistema de ayuda visual para apoyar aprendizaje de fonemas españoles

Ingeniería Electrónica, Automática y Comunicaciones

Watkins, Francisco¹  San Juan, Enrique¹  Kaschel, Héctor¹ 

关键词: digital signal rocessing;    linear predictive coding (LPC);    autocorrelation;    voice analysis;    formants;    mel-frequency cepstral coefficients (MFCCs).;   

DOI  :  

来源: Instituto Superior Politecnico "Jose Antonio Echeverria"

【 摘 要 】

This paper aims to show a computer system with the ability to support the learning of the pronunciation of a particular set of phonemes Spanish, aimed at people who are deaf or hard of hearing, which result in speech disorders. The ability to «see what it says,» can be very useful as a method for the implementation and speech rehabilitation. Display immediately, using a graph, acoustic profiles of the main parameters of the speech signal and associate them with images that represent what this has proved very stimulating an additional alternative in the field of Pathology. The system operation is based on the extraction and comparison of key parameters of the speech signals, among which one can mention the LPC (Linear Predictive Coding), Formant and Mel-frequency cepstral coefficients (MFCCs). It is expected that the system constitutes a tool to support the rehabilitation of speech disorders, replacing the auditory feedback channel for visual feedback channel. Namely, by graphical and acoustical profiles main characteristic parameters of the speech signal, with images and progress indicators, allow to set an important additional alternative tool for rehabilitation of speech disorders.

【 授权许可】

Unknown