Bibliographic citations
Briceño, C., (2012). Diseño e implementación de un sistema de reconocimiento de palabras en un FPGA basado en el algoritmo del LPC [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/2228
Briceño, C., Diseño e implementación de un sistema de reconocimiento de palabras en un FPGA basado en el algoritmo del LPC [Tesis]. : Universidad Nacional de Ingeniería; 2012. http://hdl.handle.net/20.500.14076/2228
@mastersthesis{renati/705525,
title = "Diseño e implementación de un sistema de reconocimiento de palabras en un FPGA basado en el algoritmo del LPC",
author = "Briceño Aranda, César Alberto",
publisher = "Universidad Nacional de Ingeniería",
year = "2012"
}
For several decades, humans have tried to communicate with machines in the most spontaneous way, through speech. The difficulties in getting machines to understand human speech is the result of many aspects: variability in the tone of voice, speed of pronunciation, imminent presence of noise, regional accents, distortion of language, etc. The most sophisticated speech recognition based on computer algorithms were implemented in software; therefore they are limited in processing speed. Then, new solutions arise, such as those based on hardware, using programmable logic devices such as FPGA (Field Programmable Gate Array), whose use is spreading in many aspects of daily life where you need speed and high capacity of processing, such as systems for human voice recognition. This document shows the process of designing and implementing a digital system capable of recognizing the human voice by interpreting the meaning of several words of one speaker in particular, as a solution implemented in hardware. The system takes advantage of the processing speed feature of the FPGA device, and the flexibility of the hardware description language used to program the device for generating digital circuits including embedded processors, which are programmed in C or C++, language used to achieve the implementation of sequential pattern recognition algorithms that are designed for computer systems. The project used an audio codec for data voice acquisition, LPC algorithm for voice coding, and a multilayer perceptron artificial neural network for voice-coded pattern recognition. Key words: Speech Recognition, FPGA, Artificial Neural Networks, C code, LPC.
This item is licensed under a Creative Commons License