Inventor · Los Angeles, CA, US

Mehrez Souden

19Patents
3h-index
27Co-inventors
56Inventor score

Filing activity: Jan 30, 2013 → Oct 3, 2023

Most-cited inventions

PatentTitleAreaCited byStatus
US11532306B2 Detecting a trigger of a digital assistant Electricity 25 Active
US10798511B1 Processing of audio signals for spatial audio Electricity 9 Active
US10403299B2 Multi-channel speech signal enhancement for robust voice trigger detection and automatic speech recognition Electricity 3 Active
US11341988B1 Hybrid learning-based and statistical processing techniques for voice activity detection Physics 3 Active
US11508388B1 Microphone array based deep learning for time-domain speech signal extraction Physics 3 Active
US10546593B2 Deep learning driven multi-channel filtering for speech enhancement Electricity 3 Active
US12141347B1 Machine learning and user driven selective hearing Electricity 2 Active
US10334357B2 Machine learning based sound field analysis Electricity 2 Active
US11514928B2 Spatially informed audio signal processing for user speech Physics 1 Active
US11546692B1 Audio renderer based on audiovisual information Electricity 1 Active
US11849291B2 Spatially informed acoustic echo cancelation Electricity 0 Active
US9754608B2 Noise estimation apparatus, noise estimation method, noise estimation program, and recording medium Physics 0 Active
US12283289B2 Separating and rendering voice and ambience signals by offsetting impact of device movements Electricity 0 Active
US12272370B2 Audio source separation for audio devices Electricity 0 Active
US11810588B2 Audio source separation for audio devices Electricity 0 Active
US11490218B1 Time domain neural networks for spatial audio reproduction Electricity 0 Active
US11996114B2 End-to-end time-domain multitask learning for ML-based speech enhancement Physics 0 Active
US11222652B2 Learning-based distance estimation Electricity 0 Active
US12010490B1 Audio renderer based on audiovisual information Electricity 0 Active

Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.