Publication: Síntesis de voz y reconocimiento del habla : implementación en el robot HOAP-3
Loading...
Identifiers
Publication date
2011-11
Defense date
2011-11-14
Authors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Impact
Export
Abstract
La interacción Humano-Robot es una parte casi tan importante en la creación de los robots como el diseño físico o la programación asociada. Por esta razón surge la necesidad de hacer de esta interacción lo más sencilla y amistosa posible. Desde la antigüedad, la forma de comunicación utilizada por el ser humano, ha sido mediante sonidos. Estos sonidos se han convertido a lo largo de los años en un sistema complejo con reglas, lógica y palabras para describir el mundo que rodea a los humanos: “EL LENGUAJE”. Por esta razón, la manera más sencilla de comunicación es mediante la voz. Sin necesidad de una interfaz complicada o un teclado y ratón, la voz se convierte en la herramienta principal de comunicación entre los humanos y los robots. Gracias al avance tecnológico, hoy en día el ser humano es capaz de dar órdenes a los robots mediante la voz y hacer que los robots se comuniquen mediante la síntesis de voz. La síntesis de voz consiste en la creación de ondas de sonido artificiales semejantes al habla humana. El reconocimiento del habla es un sistema capaz de transcribir un mensaje de voz en texto independientemente del hablante. Se basa en la comparación con un modelo acústico recogido en una base de datos. En este proyecto se ha conseguido la implementación de ambos sistemas (reconocimiento del habla y síntesis de voz) en el robot humanoide HOAP-3. ___________________________________________________________________________________________________________________________
Human-Robot Interaction is almost as important part in developing robots like the physical design or programming associated. For this reason there is a need to make this interaction as simple and user friendly as possible. Since ancient times, the simplest form of communication used by humans has been using sounds. These sounds have become over the years into a complex system with rules, logic and words to describe the world around humans, "LANGUAGE". For this reason, the easiest way of communication is by voice. No need for a complicated interface or a keyboard and mouse, the voice becomes the main tool of communication between humans and robots. Thanks to technological advances, today the human being is able to give commands to the robot by voice and make the robots communicate using speech synthesis. Speech synthesis (Text-to-speech) is the creation of artificial sound waves similar to human speech. It is based on the concatenation of sound units to build the voice. Speech recognition is a system capable of transcribing a voice message in text independently of the speaker. It is based on comparison with an acoustic model collected in database. This project has achieved the implementation of both systems (speech recognition and speech synthesis) in the humanoid robot HOAP-3.
Human-Robot Interaction is almost as important part in developing robots like the physical design or programming associated. For this reason there is a need to make this interaction as simple and user friendly as possible. Since ancient times, the simplest form of communication used by humans has been using sounds. These sounds have become over the years into a complex system with rules, logic and words to describe the world around humans, "LANGUAGE". For this reason, the easiest way of communication is by voice. No need for a complicated interface or a keyboard and mouse, the voice becomes the main tool of communication between humans and robots. Thanks to technological advances, today the human being is able to give commands to the robot by voice and make the robots communicate using speech synthesis. Speech synthesis (Text-to-speech) is the creation of artificial sound waves similar to human speech. It is based on the concatenation of sound units to build the voice. Speech recognition is a system capable of transcribing a voice message in text independently of the speaker. It is based on comparison with an acoustic model collected in database. This project has achieved the implementation of both systems (speech recognition and speech synthesis) in the humanoid robot HOAP-3.
Description
Keywords
Reconocimiento de voz, Robótica, Interacción hombre-máquina, Robots humanoides