Waleed Azmy | Cairo University (original) (raw)

Waleed Azmy

Uploads

Papers by Waleed Azmy

Research paper thumbnail of The Creation of Emotional Effects For An Arabic Speech Synthesis System

Having emotional effects in a speech synthesis system is an important requirement for many applic... more Having emotional effects in a speech synthesis system is an important requirement for many applications that require expressive synthesis styles. In this work we introduce the efforts done to build a unit selection based Arabic speech synthesis voice with emotional effects. Three emotional sates were covered; normal, sad and questions. Pitchmarking enhancements have been carried out for Arabic voice building for more accurate unit concatenation. The Expressive information was employed in the proposed target cost to produce more natural and emotive synthetic speech. The system is evaluated according to the naturalness and emotiveness of the produced speech. The system evaluations show significant increase in the naturalness and emotiveness scores.

Research paper thumbnail of Content-Based Recommendation System Using Search Engine

Finding information on a large web site can be a difficult and time-consuming process. Recommende... more Finding information on a large web site can be a difficult and time-consuming process. Recommender systems can help users find information by providing them with personalized suggestions. In this paper we introduce content-based recommender system TayaRecSys that suggests new stories for users in Yahki.com website. The system uses the same infrastructure of a simple search application. It is based on Lucene framework search engine. While in this work, our main focus remains on story recommendations, the proposed methods are quite general, and may apply to a wide variety of applications especially with rich text contents. The test results seem to indicate that the suggestions that TayaRecSys makes is mostly relevant.

Research paper thumbnail of Arabic Unit Selection Emotional Speech Synthesis using Blending Data Approach

International Journal of Computer Applications, 2013

Research paper thumbnail of MLP, Gaussian Processes and Negative Correlation Learning for Time Series Prediction

Multiple Classifier Systems, Jan 1, 2009

... This model was among the winning models in two major international forecasting competition th... more ... This model was among the winning models in two major international forecasting competition the NN3 and ... Future Work In this work we investigate efficient machine learning techniques and ensemble methods for time ... The application we focus on is forecasting tourist arrivals. ...

Research paper thumbnail of Forecast Combination Strategies for Handling Structural Breaks for Time Series Forecasting

Multiple Classifier Systems, Jan 1, 2010

Time-series forecasting is an important research and application area. Much effort has been devot... more Time-series forecasting is an important research and application area. Much effort has been devoted over the past decades to develop and improve the time series forecasting models based on statistical and machine learning techniques. Forecast combination is a well-...

Research paper thumbnail of The Creation of Emotional Effects For An Arabic Speech Synthesis System

Having emotional effects in a speech synthesis system is an important requirement for many applic... more Having emotional effects in a speech synthesis system is an important requirement for many applications that require expressive synthesis styles. In this work we introduce the efforts done to build a unit selection based Arabic speech synthesis voice with emotional effects. Three emotional sates were covered; normal, sad and questions. Pitchmarking enhancements have been carried out for Arabic voice building for more accurate unit concatenation. The Expressive information was employed in the proposed target cost to produce more natural and emotive synthetic speech. The system is evaluated according to the naturalness and emotiveness of the produced speech. The system evaluations show significant increase in the naturalness and emotiveness scores.

Research paper thumbnail of Content-Based Recommendation System Using Search Engine

Finding information on a large web site can be a difficult and time-consuming process. Recommende... more Finding information on a large web site can be a difficult and time-consuming process. Recommender systems can help users find information by providing them with personalized suggestions. In this paper we introduce content-based recommender system TayaRecSys that suggests new stories for users in Yahki.com website. The system uses the same infrastructure of a simple search application. It is based on Lucene framework search engine. While in this work, our main focus remains on story recommendations, the proposed methods are quite general, and may apply to a wide variety of applications especially with rich text contents. The test results seem to indicate that the suggestions that TayaRecSys makes is mostly relevant.

Research paper thumbnail of Arabic Unit Selection Emotional Speech Synthesis using Blending Data Approach

International Journal of Computer Applications, 2013

Research paper thumbnail of MLP, Gaussian Processes and Negative Correlation Learning for Time Series Prediction

Multiple Classifier Systems, Jan 1, 2009

... This model was among the winning models in two major international forecasting competition th... more ... This model was among the winning models in two major international forecasting competition the NN3 and ... Future Work In this work we investigate efficient machine learning techniques and ensemble methods for time ... The application we focus on is forecasting tourist arrivals. ...

Research paper thumbnail of Forecast Combination Strategies for Handling Structural Breaks for Time Series Forecasting

Multiple Classifier Systems, Jan 1, 2010

Time-series forecasting is an important research and application area. Much effort has been devot... more Time-series forecasting is an important research and application area. Much effort has been devoted over the past decades to develop and improve the time series forecasting models based on statistical and machine learning techniques. Forecast combination is a well-...

Log In