Jersey City, NJ, United States of America

Pedro Jose Moreno Mengibar

This inventor holds 118 USPTO granted patents and 27 published patent applications and 31 EPO patents, primarily in Automatic Speech Recognition (CPC class G10L15-197). Top assignees: Google Inc., Other. Active years: 2013-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Pedro Jose Moreno Mengibar. Retrieved from https://idiyas.com/inventor/pedro-jose-moreno-mengibar

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 118 

% Patents Active = 96.6

 

Average Co-Inventor Count = 2.6

ph-index = 15

Forward Citations = 1,113(Granted Patents)

Forward Citations (Not Self Cited) = 988(Dec 10, 2025)


Inventors with similar research interests:

🔬 See more inventors

Location History:

  • New York, NY (US) (2013)
  • Mountain View, CA (US) (2015)
  • Jersey City, CA (US) (2018)
  • Jersey City, NY (US) (2020)
  • Jersey City, NJ (US) (2013 - 2024)

Company Filing History:


Years Active: 2013-2026

Loading Chart...
Loading Chart...
Areas of Expertise:
Speech Recognition
Dynamic Language Models
Contextual Biasing
Multilingual Models
Automated Calling System
Voice Recognition
Attention Mechanisms
Speech Conversion
Dialog States
Self-Supervised Learning
Hotwording
Language Model Modulation
118 patents (USPTO):Explore Patents

Title: Innovations and Contributions of Pedro Jose Moreno Mengibar

Introduction

Pedro Jose Moreno Mengibar, located in Jersey City, NJ, is a prominent inventor with an impressive portfolio of 106 patents. His work primarily focuses on advancing technology in the fields of automatic speech recognition (ASR) and speech conversion, demonstrating noteworthy contributions to the realm of artificial intelligence.

Latest Patents

Among his latest patents, one of significant interest is titled "Advancing the use of text and speech in ASR pretraining with consistency and contrastive losses." This method involves receiving various types of training data, including unspoken text utterances, un-transcribed non-synthetic speech utterances, and transcribed non-synthetic speech utterances. Importantly, each unspoken text utterance is not paired with any corresponding spoken utterance, contributing to the unique framework of ASR. The method incorporates generating synthetic speech representations for unspoken textual utterances using a text-to-speech model and pre-training an audio encoder to better understand shared speech and text representations.

Another remarkable patent is the "Conformer-based speech conversion model." This method focuses on receiving an input spectrogram related to an utterance and processing it through an encoder comprising a stack of self-attention blocks. The output from this encoder is an encoded spectrogram, which is then used to synthesize a speech representation of the utterance via a dedicated spectrogram decoder. Collectively, these innovations exemplify his expertise in leveraging deep learning methodologies for speech processing.

Career Highlights

Pedro has had a distinguished career, with notable tenure at Google Inc., where he applied his skills in developing cutting-edge speech technologies. His professional experience underscores a dedication to advancing the capabilities of speech recognition systems, making him a key player in the tech landscape.

Collaborations

Throughout his career, Pedro has collaborated with several talented individuals, including Petar Aleksic and Fadi Biadsy. These collaborations have undoubtedly enriched his innovative endeavors and contributed to the success of his patented technologies.

Conclusion

Pedro Jose Moreno Mengibar’s innovative spirit and technical expertise, evidenced by his 106 patents, mark him as a leading figure in the field of speech technology. His recent patents push the boundaries of what is possible in speech recognition and conversion, paving the way for future advancements that may profoundly impact how technology interacts with human speech.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…