Videos with tag natural
Results 1-8 of 8
 
18:37
18:37
18:37

Increasing the Value of Video Content: Making Use of the Audio Component

An introduction to the concept of indexing video based on audio content is review through research on speech-to-text recognition, similar systems and a robust and accurate speech-to-text tool, which reduced the training time required to complete the experiment and suggests a method to improve on prior research. The paper walks through the methods to design such a system, incorporates each component showing timed results along with critical analysis of the components and prior work. The paper examines Deep Neural Network combined with Hidden Markov Models and shows substantial improvements against previous and recorded experiments and finally, shows what a system of this nature might look and perform like as a proof of concept system and offers an initial look into a method to use audio information extracted from a video a leap directly to points in the video, demonstrating the potential of content indexing.

Added: 655 days ago by mbannan

Views: 202 | Comments: 0 | Not yet rated

 
15:32
15:32
15:32

Environmental Noise and Automatic Speech Recognition

In order to make human computer interaction more users friendly and natural, it is important to minimize the speech recognition errors. The automatic speech recognition technology can be greatly improved by adding the knowledge of current environment, user's emotional state, canceling the noise in the environment and overcoming the inter-speaker speech variations. A great amount of work is already performed in other areas such as Natural Language Processing and to reduce the ASR errors. In this research paper we will answer the question: “Is the level of noise in the surrounding environment related to the performance of Automatic Speech Recognition technology in Human - Computer Spoken Dialogues”?

Main objective of this report is to identify the relationship of the environmental noise on the performance of Automatic Speech Recognition Technology.

Added: 1025 days ago by nawaz5566

Views: 672 | Comments: 0 | Not yet rated

 
10:03
10:03
10:03

Systematic Mapping of the Model Driven Architecture

There are many systematic mappings and systematic reviews of literature surrounding any given topic in software engineering however none that attempt to map the evolution of certain topics overtime in a systematic way. Rather than attempt to formulate an opinion on the evolution of model driven architecture or frameworks relating to software engineering that are model-driven this exploratory paper attempts to systematically map the trend of subjects relating to the model driven architecture using a systematic approach combining some readily available software libraries by employing natural language processing.[1] and keywording techniques.

Added: 1388 days ago by rkvirani

Views: 666 | Comments: 0 | Not yet rated

 
00:27.72
00:27.72
00:27.72

Automarking Web Service-30-second intro

The research team has leveraged the research conducted in recent Natural Language Processing studies and designed a computerized marking mechanism for marking students' short-answers based on the semantic meaning of those answers. An open access Automarking web service has been developed to provide short-answer marking service for everyone. Automarking service is also a self-sustainable complete online exam system which schools and teachers can use it to create their own item bank and tests, to manual and to auto mark students' answer sheets, and to review the results from various perspectives such as test, item, and individual student. Automarking web service is a completely open-access online exam system and marking service which realized the computerized short-answer marking mechanism proposed by the research team earlier. Not only schools, teachers, and students can benefit from the computerized marking service, but also developers and companies of computer-based test providers can integrate the mechanism into existing their CBT products and services, by communicating to Automarking web service with SOAP Envelope in XML. The component-based architecture selected seems appropriate as the ultimate goal of the system would be to support existing Learning Management Systems. As such, only those components required could be 'plugged into' an existing LMS to make the LMS capable of marking student responses of open-ended questions.

Added: 1455 days ago by maiga

Views: 9851 | Comments: 0 | Not yet rated

 
12:03
12:03
12:03

Automarking Web Service

The research team has leveraged the research conducted in recent Natural Language Processing studies and designed a computerized marking mechanism for marking students' short-answers based on the semantic meaning of those answers. An open access Automarking web service has been developed to provide short-answer marking service for everyone. Automarking service is also a self-sustainable complete online exam system which schools and teachers can use it to create their own item bank and tests, to manual and to auto mark students' answer sheets, and to review the results from various perspectives such as test, item, and individual student. Automarking web service is a completely open-access online exam system and marking service which realized the computerized short-answer marking mechanism proposed by the research team earlier. Not only schools, teachers, and students can benefit from the computerized marking service, but also developers and companies of computer-based test providers can integrate the mechanism into existing their CBT products and services, by communicating to Automarking web service with SOAP Envelope in XML. The component-based architecture selected seems appropriate as the ultimate goal of the system would be to support existing Learning Management Systems. As such, only those components required could be 'plugged into' an existing LMS to make the LMS capable of marking student responses of open-ended questions.

Added: 1455 days ago by maiga

Views: 39469 | Comments: 0 | Not yet rated

 
22:46
22:46
22:46

A Personal Virtual Assistant

From the inception of the very first computer system we realized the future potentials and possibilities when we combine a computational computer system with natural human behaviors. One of the human behaviors that scientists wish to integrate with a computer system is natural language (NL). “Natural language is any language which arises in an unpremeditated fashion as the result of the innate facility for language possessed by the human intellect” [1]. NL is natural to human being and it is being use by you and I on a daily basis for communication and interaction. NL actions majority comprised of speaking, signaling, and writing. It is through these actions that allow human being to effectively communicate with one another. With the emergence of natural language interface (NLI) type of systems we now have a mechanism that helps us effectively and efficiently communicate and interact with computer systems. Natural language based systems can tap into many different types of data sources available on the Internet to retrieve relevant information for end users. These systems can be thought of as a separate silo of mashable type of systems whereby natural language is the primary driver for their existence. With this concept in mind this paper will explore a number of ways to integrate a variety of application programming interface (API) to build an NLI system as well introducing some of the potential challenges that one may face.

Added: 1513 days ago by trancaohuy

Views: 742 | Comments: 0 | Not yet rated

 
19:56
19:56
19:56

Generating natural language specifications from UML class

Meziane, F., Athanasakis, N., & Ananiadou, S. (2008). Generating Natural Language specifications from UML class diagrams. Requirements Engineering, 13(1), 1-18. doi: 10.1007/s00766-007-0054-0. http://dx.doi.org/10.1007/s00766-007-0054-0

Added: 3293 days ago by JamesRonholm

Views: 576 | Comments: 0 | star star star star star star star star star star 

 
19:24
19:24
19:24

Text mining and software engineering

Witte, R., Li, Y., Zhang, Y. and Rilling, J.: ‘Text mining and software engineering: an integrated source code and document analysis approach’. IET Softw., 2008, 2(1), pp. 3 – 16. Retrieved September 2008 from http://0-ieeexplore.ieee.org.aupac.lib.athabascau.ca/iel5/4124007/4460888/04460890.pdf?tp=&arnumber=4460890&isnumber=4460888. doi: 10.1049/iet-sen:20070110

Added: 3293 days ago by Laurie

Views: 547 | Comments: 0 | star star star star star star star star star star