RESUMO
The development of speech is a process by which the neural network structure of brain perceives, decodes, represents, and outputs input signals. At the cortical level, the classic model of speech processing proposed two pathways leading from the auditory system: the ventral stream and the dorsal stream.The ventral stream, which involved structures in the superior and middle portions of the temporal lobe, was involved in processing speech signals for perception and comprehension (speech recognition). The dorsal stream, which involved structures in the posterior frontal lobe and the posterior dorsal-most aspect of the temporal lobe and parietal operculum, had an auditory-motor integration function and was essential for speech production. However, the importance of auditory processing in central auditory nervous system(CANS) from the vestibulocochlear nerve to the auditory cortex for speech development had not been clarified.This review aims to discuss the roles and related mechanisms of the CANS in speech recognition and generation from the perspective of cognitive neuroscience, so as to provide new ideas for clinical monitoring and long-term prediction of auditory processing and speech development, and for diagnosis and treatment of clinical speech-related diseases.
RESUMO
A medida que las tecnologías de procesamiento del habla continúan mejorando, gradualmente nos acercamos al viejo sueño de crear una máquina que hable. Los actuales sistemas interactivos de diálogo hablado permiten que los usuarios realicen tareas simples, tales como transacciones bancarias y reservas en hoteles, mediante la interacción verbal. Pese a ser relativamente exitosas, estas conversaciones humano-computadora aún tienen un largo camino para recorrer en cuanto a su naturalidad: estos sistemas tienden a ser descriptos por los usuarios como extraños o incluso intimidantes. Entre las razones principales para esta falta de naturalidad, figura el modelado imperfecto de la variación prosódica, o cómo algunas propiedades del habla (tales como la entonación, la intensidad o el ritmo) cambian en las expresiones verbales. Los sistemas actuales todavía son incapaces de manejar estas características en forma correcta, tanto al entender el habla del usuario como para producir respuestas sintetizadas. La variación prosódica es extremadamente compleja en el habla espontánea, y se sabe que la afectan varios niveles de representación lingüística (léxica, sintáctica, semántica y pragmática). En el presente artículo, enfocamos nuestra atención en una dimensión particular de variación prosódica, conocida como mimetización entre interlocutores, que consiste en la alineación automática de características del habla entre los participantes de un diálogo. Tras un repaso general de la literatura de estos temas, describimos un proyecto de investigación en curso que busca modelar la mimetización prosódica en diálogos.
As speech processing technologies continue to improve, the old dream of creating a machine that talks gradually becomes real. The present interactive speech systems enable users to perform simple tasks such as banking transactions and hotel reservations,through verbal interaction. Despite being relatively successful, these human-computer conversations still have a long way to go regarding their naturalness: these systems tend to be described as odd or even intimidating by users. Among the main reasons for this lack of naturalness, is the flawed modeling of prosodic variation or the way some properties of speech (such as intonation, intensity and rhythm) change in verbalexpressions. Current systems are still unable to handle these features correctly, both to understand the speech of the user as to produce synthesized responses. Prosodic variation is extremely complex in spontaneous speech, and it is well known that it´saffected by several levels of linguistic representation (lexical, syntactic, semantic and pragmatic). The present article focuses on a specific dimension of prosodic variation, known as mimetization between interlocutors, which consists in the automatic alignment of speech features between the participants of a dialogue. After a general overview of the literature on these subjects, a research project in process that seeks to model the prosodic mimetization in dialogues is described.