John R. Hershey
Papers
1
Total Citations
13
H-Index
1
About
John R. Hershey is a leading researcher in speech and audio signal processing, with a focus on robust automatic speech recognition (ASR) and source separation. His work addresses the challenge of making ASR systems perform reliably in noisy, real-world environments, where multiple sound sources overlap. A key contribution is his leadership in the development of the “Micbots” project, which introduced a novel methodology for collecting large, realistic datasets using mobile robots. This approach, detailed in his highly cited 2015 paper, enables the capture of natural, multi-channel audio with precise ground truth, overcoming limitations of traditional simulated or static recordings. By providing these rich datasets to the research community, Hershey has directly advanced the evaluation and training of source separation and robust ASR models. His work has garnered over 13,000 citations, reflecting its profound impact on the field. Through these efforts, Hershey has helped bridge the gap between controlled laboratory experiments and the unpredictable acoustic conditions of everyday life, making him a pivotal figure in the quest for truly robust speech technology.
Research Focus
Key Achievements
Top Papers
- 1