James Glass

Massachusetts Institute of Technology

Papers

4

Total Citations

172

H-Index

4

About

James Glass is a leading figure in speech recognition and human-robot interaction, with a career focused on making machines understand human speech in real-world, noisy environments. His work bridges the gap between laboratory accuracy and practical deployment, particularly for mobile robots and wearable devices. Glass pioneered the use of deep-neural-network acoustic models for scalable speech recognition, as demonstrated in his highly cited 2017 paper (90 citations) on hardware-accelerated ASR with voice-activated power gating, a critical innovation for power-constrained applications. He is also renowned for developing situationally aware voice interfaces for large robots, such as a multi-ton robotic forklift that operates safely alongside people in unstructured outdoor settings (47 citations). His contributions include robust voice activity detection using harmonicity and modulation frequency (20 citations), enabling reliable speech capture in low-SNR environments. Glass’s work on spoken command systems for mobile robots in outdoor military depots (15 citations) showcases his commitment to deploying speech technology in challenging, human-occupied spaces. With a career spanning foundational ASR algorithms to full-stack robotic systems, Glass has profoundly influenced how we interact with machines through voice.

Research Focus

Key Achievements

4
H-Index
4
Papers
172
Total Citations
43
Avg Citations/Paper
🏆 Most Cited Paper
14.4 A scalable speech recognizer with deep-neural-network acoustic models and voice-activated power gating
90 citations · 2017
📈 Most Prolific Year: 2017 (1 Papers)
🤝 Key Collaborators: 18
🏛 Institutions: Massachusetts Institute of Technology

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago