SRI International

SRI International - American Nonprofit Research Institute

Bios

Horacio Franco

Chief Scientist, Speech Technology and Research Laboratory (STAR)

Horacio Franco is chief scientist in the Speech Technology and Research Laboratory at SRI International. Since joining SRI in 1990, he has contributed in the areas of  acoustic modeling, hybrid Hidden Markov Model (HMM)-neural net speech recognition approaches, speech recognizer architectures, speech technology for language learning, noise-robust speech recognition, and speech-to-speech translation systems.

He has co-authored more 70 papers, as well as several communications and book chapters. Franco is co-author of several U.S. patents in aspects of speech recognition, speech technology for language learning, and noise-robust speech recognition. He has also been active in leading efforts to develop and deploy SRI’s speech technology in different areas of government and commercial use.

Franco received his engineer’s degree in electronics in 1978, and his doctor in engineering degree in 1996, both from the University of Buenos Aires.

Recent publications

more +
Wideband Spectral Monitoring Using Deep Learning (7/22/2020) - We present a system to perform spectral monitoring of a wide band of 666.5 MHz, located within a range of…
Voices Obscured in Complex Environmental Settings (VOiCES) corpus (6/1/2018) - This paper introduces the Voices Obscured in Complex Environmental Settings (VOiCES) corpus, a freely available dataset under Creative Commons BY…
Tackling Unseen Acoustic Conditions in Query-by-Example Search Using Time and Frequency Convolution for Multilingual Deep Bottleneck Features (12/1/2017) - Standard keyword spotting based on Automatic Speech Recognition (ASR) cannot be used on low- and no-resource languages due to lack…
Noise-robust Exemplar Matching for Rescoring Query-by-Example Search (12/1/2017) - This paper describes a two-step approach for keyword spotting task in which a query-by-example (QbE) search is followed by noise…
Leveraging Deep Neural Network Activation Entropy to Cope with Unseen Data in Speech Recognition (8/1/2017) - Unseen data conditions can inflict serious performance degradation on systems relying on supervised machine learning algorithms.  Because data can often…
Speech recognition in unseen and noisy channel conditions (3/1/2017) - Speech recognition in varying background conditions is a challenging problem. Acoustic condition mismatch between training and evaluation data can significantly…