Seyed Abolfazl Mortazavi

As a member of the Speech Team, I contribute to the development and enhancement of cutting-edge speech technologies. My responsibilities include:

• Text-to-Speech (TTS): Developing and fine-tuning neural network models for natural-sounding Persian speech synthesis, focusing on improving prosody, intonation, and overall voice quality. This involves data preprocessing, model training, and evaluation.
• Automatic Speech Recognition (ASR): Working on training and evaluating ASR models to enhance accuracy and robustness for the Persian language, particularly in noisy call center environments and across diverse accents. This includes acoustic and language model development.
• Call Center Quality Control: Designing and implementing AI-driven solutions to analyze call recordings for quality assurance. This involves leveraging ASR and NLP techniques for sentiment analysis, keyword spotting, topic modeling, and automated compliance checking to provide actionable insights for performance improvement.

Collaborating with cross-functional teams to integrate these speech technologies into various Shatel products and services, and utilizing Linux environments for development, testing, and deployment of AI models.

Technologies Used:

Working on diverse programming projects, specializing in Signal Processing, Artificial Intelligence, and Image Processing.

Technologies Used:

As a Research Assistant under the supervision of Dr. Farnaz Ghasemi, I played a key role in the lab's research on BCI systems and complex brain signal analysis. My responsibilities expanded to include:

• Student Supervision: Mentored and supervised several BSc and MSc students, providing technical guidance and support for their research, experimental design, and thesis projects related to EEG and BCI.
• Advanced Signal Processing: Implemented and applied Visibility Graph (VG) techniques to analyze non-linear dynamics and connectivity patterns in EEG time-series data.
• BCI & EEG Research: Conducted in-depth research on Brain-Computer Interface (BCI) systems, focusing on robust feature extraction and developing novel metrics to evaluate EEG signal quality.
• Statistical Analysis: Performed comprehensive statistical analysis and hypothesis testing on large-scale EEG signal datasets to validate research findings.

Technologies Used:

Conducted research on BCI systems and EEG signals under the supervision of Dr. Farnaz Ghasemi. Key responsibilities included developing metrics to evaluate EEG signal quality, analyzing statistical parameters of EEG signals, and conducting statistical tests for EEG signal datasets.

Technologies Used:

Worked on signal processing tasks, focusing on ECG and PPG signal analysis and feature extraction for machine learning applications. Major contributions included developing algorithms for calibrating blood pressure devices and creating methods to remove EMG noise from signals.

Technologies Used: