Samuele Cornell

Samuele Cornell

Post-Doctoral Research Associate

Language Technologies Institute
Carnegie Mellon University

About Me

I am a Post-Doctoral Research Associate at Carnegie Mellon University's Language Technologies Institute, working on speech and audio processing with machine learning. My research focuses on developing robust systems for real-world scenarios involving overlapping speech, noise, and challenging acoustic conditions.

Research Interests

Speech Separation

Developing advanced models like SepFormer for separating overlapping speech signals in challenging acoustic environments.

Speaker Diarization

Creating systems that can identify "who spoke when" in multi-speaker audio recordings.

Speech Enhancement

Improving speech quality and intelligibility by removing noise and reverberation from audio signals.

Automatic Speech Recognition

Building robust ASR systems that can accurately transcribe speech in diverse conditions.

Machine Listening

Sound event detection and acoustic scene classification for computational analysis of environmental audio, including contributions to DCASE challenges.

Notable Projects & Contributions

SpeechBrain

Co-author of this widely-used PyTorch-based speech toolkit. One of the most influential open-source projects in speech processing with 884+ citations.

View on GitHub

Asteroid

Co-author of this PyTorch-based audio source separation toolkit, enabling researchers worldwide to develop advanced speech separation systems.

View on GitHub

ESPnet-SE++

Co-author of speech enhancement extensions to ESPnet, providing state-of-the-art tools for robust speech processing.

View on GitHub

CHiME-7 & CHiME-8 DASR

Lead organizer of these challenges pushing the field toward generalizable and robust distant meeting transcription.

Learn More

URGENT Challenge

Co-organizer of this speech enhancement challenge series focusing on universality, robustness, and generalizability.

Learn More

DCASE Task 4

Led this sound event detection task for several years, advancing machine listening research and methodology.

View on GitHub

Publications

Selected publications from my research in speech processing and machine learning. For a complete list, please visit my Google Scholar profile.

Most Cited Publications

Loading publications...

Recent Publications

Loading publications...

Contact

Institution

Carnegie Mellon University
Language Technologies Institute
Pittsburgh, PA