Ahmed M. Abdelatty Ali

  • My Research
  • Education
  • Honors
  • Experience
  • Courses
  • Skills
  • Publications


    Research:

    Auditory-Based Acoustic-Phonetic Signal Processing for Robust Automatic Recognition of Speaker-Independent Continuous Speech

    Automatic speech recognition (ASR) has been an area of intense research for more than four decades. Despite recent success in this field, the current state-of-the-art ASR systems are terribly deficient when compared to the human performance. They require training when used by a new speaker, their performance deteriorates significantly in presence of noise or any mismatch from the training environment and they have limited real vocabulary size (perplexity). These systems behave as a "Black Box" with minimal integration of acoustic-phonetic, Psychological, and auditory knowledge. Part of the deficiency in performance is attributed to the limited amount of acoustic-phonetic knowledge incorporated in those systems and the limited understanding of the different sources of acoustic-phonetic variability of speech.

    To help solve these problems, we investigate the acoustic-phonetic characteristics of the continuous speech basic building units (i.e. the phonemes). We use an auditory-based front-end processing system instead of the traditional front-end systems. The acoustic-phonetic features which prove to be rich in their information content are extracted and new algorithms are designed to use these features for phoneme recognition. Such extraction and manipulation algorithms are very simple and parallel in nature. Using this approach in the fricative and stop consonants, which are the most difficult phonemes in their recognition, achieved a recognition accuracy of 90%-95% for continuous speech spoken by multiple speakers (22) from 5 different dialect regions of American English. This represents a significant improvement over the 75%-80% rate which was achieved before using traditional techniques.

    Using these feature extraction algorithms as a front-end for a state-of-the-art system, which is typically a Hidden Markov Model (HMM) system, is expected to improve robustness in presence of adverse conditions. It will integrate more speech knowledge in the recognition process and combine both phoneme-level and word-level recognition. Moreover, implementing this front-end processing in hardware, will take advantage of the parallel nature of the processing to obtain real-time fast results.

    Here is a summary file of our results in .pdf format.


    Education

    I am a PhD candidate at the Department of Electrical Engineering, University of Pennsylvania. My advisor is Prof. Jan Van der Spiegel, and my research topic is: "Acoustic-Phonetic Signal Processing for Speech Recognition". I had my M.Sc. and B.Sc. degrees (both with a grade of Distinction with Honor) from Ain Shams University, Cairo, Egypt. My Master thesis topic was: Analog Behavioral Modeling and its use in Analog VLSI Design. My B.Sc. thesis topic was: "Optical Fiber Communication Systems: Case Study and Implementation of the Receiver Amplifier".

    Before going to college, I spent my twelve years of school in an English private school (The English Mission College). I started learning English when I was 5 years old. I also learnt French for 6 years and German for 2 years. On my graduation from High School, I was the First student (Valedictorian) over all Egypt in the General Graduation Exam (the dreaded Thanaweyia Amma). This exam is a graduation exam as well as a qualifying exam to enter the university. Its score determines which university and what major the student will join. It is a unified exam over all Egyptian schools.


    Honors:

    1. The Egyptian Ministry of Education award of Excellence (1986) for being the First student over all Egypt in the High School General Graduation Exam. (Thanaweyia Amma).

    2. The Cairo Governorate award of Excellence (1986) for being the First student over all Egypt in the High School General Graduation Exam. (Thanaweyia Amma).

    3. Ain Shams University award of Excellence (1991) for being the First student over my class for 5 consecutive years (1986-1991).

    4. The Egyptian Ministry of Education award of Excellence (1991) for being the First student over my class in the B.Sc. degree.

    5. University Fellowship from the University of Pennsylvania, 1994-1995.

    6. George Stephenson Foundation Fellowship, 1995-1996.


    Experience:

    1. January, 1995 - present:
    PhD student in the Electrical Engineering department, University of Pennsylvania. My research involves the use of Acoustic-Phonetic Feature-Based Signal Processing for Automatic Speech Recognition.

    2. 1993-1995:
    Working in developing Electrical Engineering Software CAD tools for "ANACAD EES", Germany (now part of "Mentor Graphics"). My work involved building Analog Behavioral Models for complicated circuits like DC-DC Converters, Operational Amplifiers, Comparators, A/D Converters, etc. I also designed and developed some CAD GUI software packages such as the "Automatic Model Generator".

    3. Summer of 1990:
    Summer training in "Schlumberger Wireline and Testing". My work involved studying and operating the electronic equipment used for measurements and sensing in oil wells.

    4. Summers of 1988 and 1989:
    Summer training in Siemens AG, Erlangen, Germany. I was involved in the design and development of the Automatic Control Systems used to operate state-of-the-art cement plants.

    5. 1991-1994:
    Assistant Lecturer/Research Assistant in the Electronics and Communication Engineering department, Ain Shams University, Cairo, Egypt.

    6. 1986-1991:
    Undergraduate Student in the Electrical Engineering department, Electronics and Communication section, Ain Shams University, Cairo, Egypt. (5 years Program).


    Courses:


    Skills:


    Publications:

    1. A.M.Abdelatty, H.Haddara and H.F.Ragaie, "Analog Behavioral Modeling of Artificial Neural Networks", International Conference on Microelectronics (ICM), Turkey, 1994.

    2. A.M.Abdelatty and H. Haddara, "Automatic Generation of Analog Behavioral Models", IEEE International Conference on Electronics, Circuits and Systems (ICECS), 1994.


    Number of visitors since 8/19/99 (when the counter was last reset):

    Back to my home page.


    Ahmed M. Abdelatty Ali


    (C) Copyright 1996-1998 by Ahmed M. Abdelatty Ali. All Rights Reserved.