Description: 一种简单有效的基于动态时变语音识别源码
对于大多数研究者来说,寻找能够匹配二重时间序列信号的最佳途径是很重要的,因为它有许多重要的应用需求.DTW是实现这项工作的显著技术,尤其在语音识别技术领域,在这里一个测试信号被按照参照模板拉伸或压缩,
-Searching for the best path that matches two time-series signals is the main task for many researchers, because of its importance in these applications. Dynamic Time-Warping (DTW) is one of the prominent techniques to accomplish this task, especially in speech recognition systems. DTW is a cost minimisation matching technique, in which a test signal is stretched or compressed according to a reference template.
Although there are other advanced techniques in speech recognition such as the hidden Markov modelling (HMM) and artificial neural network (ANN) techniques, the DTW is widely used in the small-scale embedded-speech recognition systems such as those embedded in cell phones. The reason for this is owing to the simplicity of the hardware implementation of the DTW engine, which makes it suitable for many mobile devices. Additionally, the training procedure in DTW is very simple and fast, as compared with the HMM and ANN rivals.
Platform: |
Size: 2658304 |
Author:宋小小 |
Hits:
Description: It s a neural segreatory that recognize people and speech with these. If a people is not recognized Rubycon ask to create a training with thi new people for a next recognition.
Rubycon use the Microsoft Speech sdk, gann neural network and other library (into package). It s only a sperimental software -It s a neural segreatory that recognize people and speech with these. If a people is not recognized Rubycon ask to create a training with thi new people for a next recognition.
Rubycon use the Microsoft Speech sdk, gann neural network and other library (into package). It s only a sperimental software .... Platform: |
Size: 54043648 |
Author:Flavio |
Hits:
Description: Isolated speech recognition using FFT for features extraction and Artificial Neural Network with Back Propagation for classificassion and recognition. Platform: |
Size: 149504 |
Author:Utis Sutisna |
Hits:
Description: 就目前三种主流的语音识别算法：动态时间规(DTW)、隐马尔科夫模型(HMM)和人工神经网络(ANN)。分析它们的原理、特点及实现过程，对 DTW 的语音识别进行实验，通过对比分析三种算法的特点，结合本文研究的实际情况，选择 DTW 作为研究的重点，提出利用遗传算法对其进行改进。
-The three mainstream speech recognition algorithms: Dynamic Time Regulations (DTW), hidden Markov model (HMM) and artificial neural network (ANN). Analysis of their principles, characteristics and implementation process, DTW speech recognition experiment, by comparing the characteristics of the three algorithms, combined with the actual situation of this study, select the DTW as the focus of the study, proposed the use of genetic algorithms to improve it. Platform: |
Size: 1090560 |
Author:lumeng |
Hits:
Description: 调信息在汉语语音识别中具有非常重要的意义。采用支持向量机对连续汉语连续语音进行声调识别实
验，首先采用基于Teager能量算子和过零率的两级判别策略对连续语音进行浊音段提取，然后建立了适合于支持向
量机分类模型的等维声调特征向量。使用6个二类SVM模型对非特定人汉语普通话的4种声调进行分类识别，与
BP神经网络相比，支持向量杌具有更高的识别率。-Tone is an essential component for word formation in Chinese languages．It plays a very important role in the
transmission of information in speech communication．We looked at using support vector machines(SVMs)for auto—
matic tone recognition in continuously spoken Mandarin．The voiced segments were detected based on Teager Energy
Operation and ZCIL Compared with BP neural network。considerable improvement was achieved by adopting 6 binary-
SVMs scheme in a speaker-independent Mandarin tone recognition system． Platform: |
Size: 316416 |
Author:沈 |
Hits:
Description: 工神经网络（Artificial Neural Network）又称连接机模型，是在现代神经学、生物学、心理学等学科研究的基础上产生的，它反映了生物神经系统处理外界事物的基本过程，是在模拟人脑神经组织的基础上发展起来的计算系统，是由大量处理单元通过广泛互联而构成的网络体系，它具有生物神经系统的基本特征，在一定程度上反映了人脑功能的若干反映，是对生物系统的某种模拟，具有大规模并行、分布式处理、自组织、自学习等优点，被广泛应用于语音分析、图像识别、数字水印、计算机视觉等很多领域，取得了许多突出的成果。最近由于人工神经网络的快速发展，它已经成为模式识别的强有力的工具。神经网络的运用展开了新的领域，解决其它模式识别不能解决的问题，其分类功能特别适合于模式识别与分类的应用。多层前向BP网络是目前应用最多的一种神经网络形式-Artificial neural network (Artificial Neural Network) connection, also known as machine model, is based on interdisciplinary research in modern neurology, biology, psychology, etc. produced on, it reflects the fundamental processes of biological neural processing of external things, is in the simulation developed on the basis of human brain tissue computing system is constituted by a large number of processing units interconnected through an extensive network system, it has the basic characteristics of biological neural systems, to a certain extent reflects the number of reflecting the human brain function is simulation of certain biological systems, with massively parallel, distributed processing, self-organizing, self-learning, etc., are widely used in many areas of speech analysis, image recognition, digital watermarking, computer vision, and achieved many outstanding achievements . Recently due to the rapid development of artificial neural networks, it has become a powerful tool fo Platform: |
Size: 100352 |
Author:沈阳阳 |
Hits:
Description: 基于BP网络的语言识别，选取民歌、古筝、摇滚和流行四类不同音乐，用BP神经网络实现对这四类音乐的有效分类。-Speech recognition based on BP network, songs, zither, four different types of rock and pop music, with BP neural network classification of these four types of music valid. Platform: |
Size: 373760 |
Author:zhanglei |
Hits:
Description: 程序中选取了民歌、古筝、摇滚和流行四类不同音乐 ，用 BP 神经网络实现对这四类音乐
的有效分类。(The program selected folk songs, zither, rock and popular four different music, with BP neural network to achieve the effective classification of these four types of music.) Platform: |
Size: 375808 |
Author:云泽99
|
Hits:
Description: Artificial intelligence (AI) is concerned with building systems that simulate intelligent
behavior. It encompasses a wide range of approaches, including those based on logic,
search, and probabilistic reasoning. Machine learning is a subset of AI that learns to
make decisions by fitting mathematical models to observed data. This area has seen
explosive growth and is now (incorrectly) almost synonymous with the term AI.
A deep neural network is one type of machine learning model, and when this model is
fitted to data, this is referred to as deep learning. At the time of writing, deep networks
are the most powerful and practical machine learning models and are often encountered
in day-to-day life. It is commonplace to translate text from another language using a
natural language processing algorithm, to search the internet for images of a particular
object using a computer vision system, or to converse with a digital assistant via a speech
recognition interface. All of these applications are powered by deep learning.
As the title suggests, this book aims to help a reader new to this field understand
the principles behind deep learning. The book is neither terribly theoretical (there are
no proofs) nor extremely practical (there is almost no code). The goal is to explain the
underlying ideas; after consuming this volume, the reader will be able to apply deep
learning to novel situations where there is no existing recipe for success.
Machine learning methods can coarsely be divided into three areas: supervised, unsupervised, and reinforcement learning. At the time of writing, the cutting-edge methods
in all three areas rely on deep learning (figure 1.1). This introductory chapter describes
these three areas at a high level, and this taxonomy is also loosely reflected in the book’s
organization. Platform: |
Size: 11646296 |
Author:ihaveap1 |
Hits:
Search - speech recognition with neural network