Pejman Mowlaee
Papers
1
Total Citations
12
H-Index
1
About
Pejman Mowlaee is a researcher whose work centers on advancing speech processing technologies, with a particular focus on single-channel speech separation and enhancement. His key contributions lie in developing unsupervised methods to disentangle overlapping speech signals from a single microphone—a challenging problem with applications in hearing aids, telecommunications, and voice-controlled systems. His most-cited paper, "Unsupervised single channel speech separation based on optimized subspace separation" (2017, 12 citations), introduces a novel approach that leverages subspace decomposition and optimization techniques to isolate individual speakers without requiring labeled training data. This work is notable for its ability to operate in real-world, noisy environments where traditional supervised methods often fail. Mowlaee’s research has been instrumental in pushing the boundaries of unsupervised learning for audio source separation, offering a computationally efficient alternative to deep learning-based approaches. His achievements reflect a commitment to solving practical, real-time speech processing challenges, making his work a valuable reference for students and researchers exploring robust, data-efficient speech separation systems.
Research Focus
Key Achievements
Top Papers
- 1